Procedural Sedation: Safe, Titrated, and Monitored

A Grand Rounds Review for Residents, Registrars and Consultants

Dr Neeraj Manikath DNB

---

Reader's note: Drug doses below are typical adult starting doses for a healthy adult of about 70 kg. They must be reduced for age, frailty, hepatic or renal impairment, and co-administered sedatives or opioids. Always check your local formulary and the product label.

---

1. Introduction: The Patient Who Did Everything "By the Book"

A 68-year-old man with moderate COPD arrives in the emergency department with an anterior shoulder dislocation. He is alert, comfortable at rest, and saturating 94% on room air. The registrar, who is competent and well-meaning, decides on sedation. She gives midazolam 5 mg and fentanyl 100 mcg in quick succession because "the shoulder is still hurting." Ninety seconds later she attempts the reduction.

The patient stops responding to voice. His chest rises, then barely rises. The pulse oximeter reads 84%, but only after the tone has dropped. The nurse calls out the number, the room goes silent, and the reduction is abandoned while somebody finds a bag-valve-mask.

He recovers fully. Nobody was harmed, yet the whole room knew how close it had been.

Nothing in that story was a knowledge gap about the drugs. The registrar knew midazolam and fentanyl. The failure was one of pharmacology in time: two synergistic respiratory depressants, stacked before either had reached peak effect, in an older patient with limited ventilatory reserve, with no early warning monitor beyond a lagging pulse oximeter.

Why this matters:

• Procedural sedation and analgesia (PSA) is among the most frequent high-risk interventions in modern hospital practice. It is performed in emergency departments, endoscopy suites, bronchoscopy rooms, cath labs, ICUs, wards, radiology and day-care units, often by clinicians whose main job is something else.
• In large series, serious adverse events (airway intervention, aspiration, cardiovascular collapse, death) are uncommon. Minor respiratory events are common, and they are the precursors of the serious ones.
• Nearly all deaths and permanent injuries from sedation share one mechanism: unrecognised or inadequately treated hypoventilation and airway obstruction.[1,2]

🎯 The central teaching of this article: Sedation is not a drug problem. It is a monitoring, timing and rescue problem. The best sedationists are not the ones who know the most drugs. They are the ones who see trouble coming one minute before it arrives.

This review is organised as a bedside talk would be, moving from principles to pearls to practical hacks, and then to decision thresholds that tell you when to carry on and when to call for help.

---

2. Pathophysiology: Only What Changes What You Do

2.1 The continuum is a trap and a map

The American Society of Anesthesiologists (ASA) defines four states: minimal sedation, moderate sedation, deep sedation, and general anaesthesia. They are a continuum, not discrete steps.[2]

State	Response	Airway	Ventilation	Cardiovascular
Minimal (anxiolysis)	Normal response to voice	Unaffected	Unaffected	Unaffected
Moderate ("conscious sedation")	Purposeful response to voice or light touch	No intervention needed	Adequate	Usually maintained
Deep	Purposeful response only after repeated or painful stimulation	Intervention may be needed	May be inadequate	Usually maintained
General anaesthesia	Unrousable	Intervention often required	Frequently inadequate	May be impaired

🪙 The rule of one level: The practical safety principle is that the clinician must be able to rescue a patient from one level deeper than intended. If you aim for moderate sedation, you must be competent and equipped to manage deep sedation. If you aim for deep sedation, you must be ready to manage general anaesthesia, including airway rescue.

2.2 The three physiological ideas that matter

(a) Depth depends on the drug and the stimulus. A patient who is comfortably sedated during a painful manipulation can become deeply sedated or apnoeic in the seconds after the stimulus stops. Examples are the moment the shoulder clunks in, the moment the bronchoscope leaves the airway, or the moment a cardioversion shock is delivered. Pain was acting as a ventilatory stimulant. When it is removed, the same dose becomes a deeper dose.

(b) Hypoventilation precedes hypoxaemia, especially on oxygen. Sedatives and opioids blunt the ventilatory response to CO₂ and relax upper airway tone (tongue, pharynx). The sequence is:

1. Reduced airway tone gives obstruction.
2. Reduced respiratory drive gives hypoventilation, with a rising PaCO₂.
3. Only later does oxygen saturation fall.

Supplemental oxygen delays desaturation while CO₂ continues to climb. A patient can be 98% saturated on 4 L/min and be profoundly hypoventilating.

(c) Synergy is supra-additive, and peak effect is delayed. Opioids combined with benzodiazepines or propofol do not add. They multiply. Each agent also has an effect-site delay, which means the time between injection and peak effect is longer than the time you are willing to wait.

Drug	Onset	Peak effect	Clinical duration
Propofol	30–60 s	1–2 min	5–15 min
Midazolam	1–3 min	3–5 min	30–60+ min
Fentanyl	1–2 min	3–5 min	30–60 min
Ketamine IV	30–60 s	1–5 min	10–20 min (dissociation); 60 min recovery
Etomidate	30–60 s	1–2 min	5–10 min
Dexmedetomidine	10–15 min	15–30 min	60+ min
Remimazolam	1–3 min	2–4 min	10–20 min

⚡ Rule of the peak: Never re-dose a drug before its peak has passed. The 90-second stack in the vignette is the most common cause of avoidable respiratory events.

2.3 Who has a short runway?

Airway collapse and desaturation come faster in:

• Older and frail patients (smaller central compartment, higher brain sensitivity, lower cardiac output). Dose reductions of 30–50% are routine.
• Obesity and obstructive sleep apnoea (low functional residual capacity gives rapid desaturation, and upper airway collapse is easy).
• COPD, pulmonary hypertension, neuromuscular disease (limited ventilatory reserve).
• Hypovolaemia, sepsis, heart failure (propofol gives vasodilatation and a profound pressure drop).
• Chronic opioid, benzodiazepine or alcohol use (variable; tolerance means larger doses, but hepatic or renal impairment can paradoxically prolong effects).
• Pregnancy (reduced FRC, higher oxygen consumption, aspiration risk).

---

3. Diagnostic Nuances: The Pre-Sedation Assessment That Separates Good from Great

"Diagnosis" in sedation means risk stratification before the first drop is drawn up. A thorough assessment takes 2–3 minutes. Skipping it takes seconds, and the consequence can be permanent.

3.1 The history: AMPLE plus three

Use AMPLE as the base:

• Allergies
• Medications (particularly sedatives, opioids, anticoagulants, CYP3A4 interactors)
• Past history, especially previous sedation or anaesthesia problems
• Last oral intake
• Events leading to the procedure

Then add the three that most clinicians forget:

1. Snoring, witnessed apnoeas, daytime sleepiness (use STOP-BANG). Undiagnosed OSA is the largest hidden risk in sedation.
2. Functional capacity and frailty. A patient who cannot climb a flight of stairs has little reserve. The Clinical Frailty Scale takes a minute and predicts sensitivity.
3. Alcohol, opioid and benzodiazepine use, which change dose requirements and predict withdrawal and agitation.

🦪 Oyster: Check the drug list for CYP3A4 inhibitors. Clarithromycin, erythromycin, azole antifungals, diltiazem, verapamil, ritonavir and some others can multiply the effect of midazolam and fentanyl several-fold. A patient on clarithromycin can effectively receive three times your intended midazolam dose. Reduce doses or choose an agent with a different clearance pathway (for example propofol or ketamine).

3.2 The airway examination

Predicting a difficult airway is imperfect, but a few minutes of structured thought catches obvious problems:

• LEMON: Look externally (beard, short neck, facial trauma), Evaluate 3-3-2 (mouth opening, mandibular space, thyromental distance), Mallampati score, Obstruction (tumour, haematoma, epiglottitis), Neck mobility.
• Ask: Can the patient lie flat? If not, deep sedation in the supine position is hazardous.
• Note prior difficult intubation, radiotherapy to the neck, or limited mouth opening.

3.3 ASA Physical Status and what it means for who sedates

ASA class	Description	Typical suitability for non-anaesthetist-led sedation
I	Healthy	Yes
II	Mild systemic disease, well controlled	Yes
III	Severe systemic disease (e.g. COPD, CKD, poorly controlled DM)	With caution and a clear plan
IV	Severe disease that is a constant threat to life	Involve anaesthesia
V	Moribund	Involve anaesthesia

3.4 Fasting: the practice that has outrun its evidence

🪙 Pearl: Aspiration during procedural sedation is rare, and fasting status has not reliably predicted it. A comprehensive systematic review found aspiration events to be vanishingly uncommon in non-intubated sedation.[3] An international multidisciplinary consensus statement concluded that, for elective procedural sedation in appropriately selected, low-risk patients, prolonged fasting is of uncertain benefit, and that urgent procedures should not be delayed solely because of recent oral intake.[4]

Use judgement, not ritual:

• Emergent, time-critical procedure: do not delay for fasting. Choose the agent and depth that minimise aspiration risk (see ketamine below).
• Elective deep sedation in a high-risk patient (gastric outlet obstruction, gastroparesis, active vomiting, bowel obstruction, pregnancy, significant obesity or reflux): follow local fasting policy and consider anaesthesia support.
• Do not apply an automatic six-to-eight-hour delay to every patient.

3.5 Baselines you must record

Before the first dose, document baseline SpO₂, heart rate, blood pressure, respiratory rate, level of consciousness, and (where possible) EtCO₂ waveform. These are your comparison points. A drop of 10 mmHg in BP or a 10 mmHg rise in EtCO₂ is only meaningful if you know where it started.

---

4. Monitoring and Preparation: The Non-Negotiables

4.1 Staffing

• A dedicated person, other than the proceduralist, whose only job is to monitor the patient and airway. The operator is busy operating.
• That person must be trained in airway management and advanced life support.
• At least one additional person to assist.

4.2 Equipment: SOAPME

⚡ Hack: SOAPME (borrowed from paediatric sedation practice and valuable for adults)

• Suction (working, tip in reach)
• Oxygen (source and a delivery device, including a bag-valve-mask)
• Airway equipment (oral and nasal airways, laryngeal mask, laryngoscope, correctly sized tubes)
• Pharmacy (sedatives, reversal agents, emergency drugs drawn up or immediately available)
• Monitors (continuous pulse oximetry, ECG, NIBP, capnography)
• Equipment (defibrillator, IV access, specialty items for the procedure)

If you cannot tick all six, you are not ready.

4.3 Capnography: the single biggest upgrade in sedation safety

🪙 Pearl: Capnography sees hypoventilation before the pulse oximeter does. Pulse oximetry detects hypoxaemia, which is a late event. Capnography detects apnoea and hypoventilation within seconds. A meta-analysis found that capnography monitoring substantially increased detection of respiratory events during procedural sedation compared with standard monitoring alone.[5]

How to use it:

• Waveform presence matters more than the absolute number. A flat line means the patient is not breathing (or the sampling line is dislodged. Check the patient first).
• Trends matter. A rise of more than about 10 mmHg from baseline, an absolute value above about 50 mmHg, or an abrupt fall in amplitude all suggest hypoventilation and demand action.
• In mouth breathers or those on high-flow oxygen, nasal sampling underestimates EtCO₂. Look at the shape and presence of the waveform rather than trusting the number.

🦪 Oyster: A normal pulse oximeter reading proves nothing about ventilation. A patient can be apnoeic on 100% oxygen for 60–90 seconds before the saturation falls. If your sole monitor is the pulse oximeter, you are watching a lagging indicator.

---

5. The Drugs: Choices, Doses, Timing and Pitfalls

5.1 The principle: match the drug to the procedure and the patient

Need	Best-suited approach
Short, painful procedure (fracture or joint reduction, abscess drainage, cardioversion with analgesia)	Ketamine, or propofol with an analgesic component
Short, non-painful but immobility-dependent procedure	Propofol, remimazolam, or midazolam
Haemodynamically fragile	Ketamine (with caution), etomidate, or titrated remimazolam
Cardioversion	Propofol (low dose) or etomidate
Prolonged, light sedation with preserved ventilation	Dexmedetomidine
Anxiolysis only	Midazolam, or non-pharmacological measures

5.2 Propofol

• Dose: 0.5–1 mg/kg IV bolus over 30–60 seconds, then 0.25–0.5 mg/kg every 1–3 minutes as needed. In older or frail patients, start at 0.25–0.5 mg/kg.
• Strengths: rapid onset and offset, antiemetic, excellent for short procedures.
• Weaknesses: no analgesia, dose-dependent respiratory depression and apnoea, vasodilation and hypotension, pain on injection.
• Track record: Large series of propofol given by trained non-anaesthetists (for example endoscopist-directed propofol) show very low rates of serious events when proper monitoring and rescue are in place.[6]

⚡ Hacks for propofol

• Pain on injection: give lidocaine 20–40 mg IV immediately before (or mixed in the syringe), or use a larger antecubital vein.
• Obese patients: dose bolus on lean or ideal body weight, not actual weight.
• Hypotension: treat with fluids and a small vasopressor dose (phenylephrine or ephedrine). Do not simply give more fluid to a patient with heart failure.

🦪 Oyster: Egg allergy is not a contraindication to propofol. Egg allergy is directed at egg-white proteins. Propofol's lipid emulsion uses egg lecithin (yolk-derived phospholipid), which does not carry the allergenic protein. Soy and peanut allergies likewise do not contraindicate propofol in standard practice.

5.3 Ketamine

• Dose (adult): 1–1.5 mg/kg IV, given over 30–60 seconds (not as a rapid push), with repeat doses of 0.5–1 mg/kg after 10–15 minutes if needed. IM dose: 4–5 mg/kg, with a possible repeat of half the dose.
• Mechanism: NMDA-receptor antagonism producing dissociation, with potent analgesia, amnesia, and relative preservation of airway reflexes and spontaneous ventilation.
• Strengths: excellent for painful procedures, haemodynamic stability in most patients, bronchodilatory effect.
• Weaknesses: emergence reactions, nausea and vomiting, hypersalivation, laryngospasm (rare), and hypertension and tachycardia.

Evidence-based guidance on ketamine dissociative sedation has been codified in an ED clinical practice guideline.[7] An individual-patient meta-analysis of paediatric data identified that rapid IV administration (under about 30 seconds), high cumulative doses, and oropharyngeal procedures increase airway and respiratory adverse events.[8] Adult data show a similar profile of adverse events, most of them minor and self-limiting.[9]

🪙 Pearl: Don't give midazolam "to prevent emergence reactions." A randomised trial in adults found that adding midazolam to ketamine did not reduce recovery agitation.[10] What does help: a quiet, low-stimulation environment, calm verbal reassurance, and avoiding unnecessary touch or talking during emergence. Treat severe emergence agitation with a small dose of midazolam or propofol after the fact.

🦪 Oysters on ketamine

• "Ketamine raises intracranial pressure" is a persistent myth that is not supported by modern evidence. It is not contraindicated in head injury when ventilation is preserved.
• "Ketamine is always safe in shock." In severely catecholamine-depleted patients, ketamine is a direct myocardial depressant, and hypotension can occur. Use reduced doses and be prepared.
• Routine atropine and ondansetron are not required. Consider antiemetic prophylaxis only in high-risk patients.
• Relative contraindications: active psychosis, uncontrolled hypertension or severe ischaemic heart disease, and raised intraocular pressure with an open-globe injury.

5.4 Ketofol (propofol plus ketamine)

The idea is to blend propofol's rapid, smooth hypnosis with ketamine's analgesia and haemodynamic support. A randomised trial in the ED compared propofol alone, 1:1 ketofol and 4:1 propofol:ketamine for deep sedation and found that the combinations did not produce a dramatic safety advantage over propofol alone in respiratory events, though recovery and some secondary outcomes differed.[11]

⚡ Practical point: Ketofol is a reasonable option where analgesia and haemodynamics are both concerns. It is not a mandatory upgrade. Titrate each component separately if you can, and do not assume that mixing eliminates the risk of apnoea.

5.5 Midazolam and fentanyl: the old workhorses and their traps

Midazolam

• Dose: 0.02–0.03 mg/kg IV (about 1–2 mg), titrated every 3–5 minutes. Older or frail patients: 0.5–1 mg increments.
• Pitfalls: slow peak, prolonged and unpredictable duration, accumulation in hepatic or renal failure, paradoxical agitation, and respiratory depression, especially combined with opioids.

Fentanyl

• Dose: 0.5–1 mcg/kg IV (typically 25–50 mcg increments in older patients), repeat at 3–5 minute intervals.
• Pitfalls: chest-wall rigidity with rapid large doses, synergy with benzodiazepines.

⚡ Hack: One drug, one dose, one peak. Give one small increment of one drug, then wait for the peak (3–5 minutes) before deciding whether to give another. When combining an opioid and a benzodiazepine, give the opioid first (analgesia is the real need), let it peak, and then give a reduced dose of the sedative.

🦪 Oyster: Midazolam does not provide analgesia. If the patient is moving because of pain, more midazolam gives a deeper, still-moving, hypoventilating patient. Treat the pain with an analgesic or a local or regional block.

5.6 Etomidate

• Dose: 0.1–0.15 mg/kg IV (about 7–10 mg for an adult).
• Strengths: haemodynamic stability, rapid onset and offset.
• Pitfalls: myoclonus (often mistaken for seizure; it is not), nausea, and adrenal suppression. After a single sedation dose the adrenal suppression is generally not clinically significant, but it matters in sepsis or repeated dosing.

5.7 Dexmedetomidine

• Dose: loading 1 mcg/kg over 10 minutes, then 0.2–1 mcg/kg/h.
• Strengths: relatively little respiratory depression, cooperative "arousable" sedation, anxiolysis and analgesia.
• Weaknesses: slow onset, bradycardia and hypotension, and a long duration that does not suit short procedures.

Dexmedetomidine has been studied in monitored anaesthesia care and is valuable for procedures needing a cooperative patient, for example awake fibreoptic intubation, and for prolonged light sedation.[12]

5.8 Reversal agents: know them, but do not rely on them

• Flumazenil: 0.2 mg IV, repeat to a maximum of about 1 mg. Its half-life is shorter than that of many benzodiazepines, so re-sedation can occur. It can precipitate seizures in chronic benzodiazepine users and in mixed overdoses.
• Naloxone: use small titrated doses (0.04–0.1 mg) to reverse respiratory depression without provoking acute pain, agitation or pulmonary oedema. Observe for renarcotisation.

🪙 Pearl: If you need a reversal agent, you have already made an error in dose, timing, or monitoring. Reversal agents are a safety net, not a strategy. After any reversal, observe for at least 2 hours.

---

6. State-of-the-Art Updates: What Has Changed Practice

6.1 Remimazolam: a "better" benzodiazepine?

Remimazolam is an ultra-short-acting benzodiazepine metabolised by tissue esterases (not hepatic CYP enzymes), with predictable offset and reversibility with flumazenil. In phase III trials for colonoscopy and bronchoscopy, it produced reliable sedation with faster recovery than midazolam and fewer haemodynamic effects than propofol.[13,14]

What this means in practice:

• It is attractive for patients at risk of hypotension and for short procedures needing quick recovery.
• Dosing is typically an initial dose of around 5 mg (lower with opioid co-administration), then small top-ups of about 2.5 mg at intervals of at least 2 minutes. Always follow the product label.
• It does not remove the need for monitoring. Respiratory depression still occurs, particularly when it is combined with opioids.

🦪 Oyster: Remimazolam's esterase metabolism makes it less dependent on hepatic function, a theoretical advantage in cirrhosis and in the elderly. It is also less likely to accumulate with repeated dosing than midazolam. However, long-term data in high-risk groups remain limited, so use it where its pharmacology offers a real advantage, not merely because it is new.

6.2 High-flow nasal oxygen (HFNO) during sedation

Standard nasal cannula or face-mask oxygen gives a modest reserve. High-flow nasal oxygen, which delivers heated, humidified oxygen at up to 60 L/min, produces a small amount of positive airway pressure and apnoeic oxygenation, and it reduces the incidence of hypoxaemia during procedures such as gastroscopy under propofol sedation.[15]

• Most valuable in obese, OSA, COPD and high-risk patients.
• Caveat: HFNO can mask hypoventilation. Without capnography, a high saturation may reassure falsely. Capnography remains essential.

6.3 Consensus guidance has consolidated

• The ASA 2018 practice guidelines for moderate procedural sedation and analgesia emphasise preprocedural assessment, continuous monitoring including capnography, availability of reversal agents, and the principle of rescue.[2]
• The ACEP clinical policy supports propofol, etomidate and ketamine for ED sedation and endorses capnography as a monitoring adjunct.[16]
• The international fasting consensus reframed fasting as a risk-based decision rather than a fixed rule.[4]

6.4 The honest summary of the evidence

Question	Where the evidence points
Is capnography worth using?	Yes. It detects respiratory events earlier.
Is propofol safe in non-anaesthetist hands?	Yes, with training, monitoring, and rescue capability.[6]
Does ketofol clearly beat propofol or ketamine alone?	No consistent advantage.[11]
Does fasting prevent aspiration in sedation?	Not convincingly demonstrated.[3,4]
Does midazolam prevent ketamine emergence reactions?	No.[10]
Does HFNO reduce hypoxaemia in sedation?	Yes, in selected patients.[15]
Is remimazolam non-inferior in efficacy with a hypotension advantage?	Promising.[13,14]

---

7. Clinical Pearls 🪙: A Collected Reference

🪙 1. Sedation depth is drug dose plus stimulus. The deepest moment is the one after the painful stimulus ends. Be most vigilant at that moment.

🪙 2. "Talking" is a monitor. In moderate sedation, a patient who responds to your voice is ventilating and protecting the airway. Ask the patient to say their name every few minutes. When the answer is slow or slurred, stop dosing.

🪙 3. Hypoxia is a late sign, hypercapnia an early one. If you wait for the alarm, the margin has already gone.

🪙 4. Most airway events are obstruction before they are apnoea. A jaw thrust or chin lift fixes the majority.

🪙 5. Pain is a respiratory stimulant. Treat pain with analgesics, local anaesthesia and regional blocks first. Sedation is the adjunct.

🪙 6. Older patients need less, and need more time. Cut the dose by a third to a half, and double your wait for peak.

🪙 7. The most dangerous drug combination is the one given in a hurry.

---

8. Oysters 🦪: Hidden Gems Most Clinicians Underappreciate

🦪 1. Local and regional anaesthesia first. A well-placed hematoma block for a distal radius fracture, or an intra-articular lidocaine injection for a shoulder, may remove the need for deep sedation in suitable patients. Fewer drugs, fewer events.

🦪 2. The capnography waveform shape reveals airway obstruction. A flattened, "shark-fin" or irregular waveform in a sedated patient suggests partial obstruction or bronchospasm before the saturation changes.

🦪 3. Chronic benzodiazepine or alcohol users may need larger doses, but they also hide a withdrawal trap. Reversing with flumazenil in this group risks seizures.

🦪 4. Cirrhosis and hepatic encephalopathy. Benzodiazepines can precipitate or worsen encephalopathy. Prefer agents with shorter, less hepatically dependent kinetics, and reduce doses substantially.

🦪 5. Renal failure. The active metabolites of midazolam can accumulate. Fentanyl is a better opioid choice than morphine, which has active, renally cleared metabolites.

🦪 6. Don't forget to ask about pregnancy in women of reproductive age before elective sedation or radiology involving sedation.

🦪 7. The post-procedure hour is not "safe." Drugs with long tails (midazolam, fentanyl) can produce late respiratory depression when the stimulation of the procedure has gone and the patient is left alone in a quiet room.

---

9. Clinical Hacks & Tips ⚡: Master Clinician Shortcuts

⚡ 1. Pre-oxygenate for 3 minutes before the first dose (a tight-fitting mask or high-flow device). This builds an oxygen reservoir in the lungs and buys time during an apnoeic event.

⚡ 2. Apnoeic oxygenation. Keep a nasal cannula at 10–15 L/min on the patient even while a mask is used for bag-mask ventilation, to continue oxygen delivery during airway manoeuvres.

⚡ 3. Draw up and label before you start. Pre-drawn, labelled syringes of the sedative, a reversal agent, and an emergency vasopressor (for example phenylephrine) remove cognitive load in a crisis. Use consistent concentrations (for example ketamine 10 mg/mL).

⚡ 4. Position deliberately. A shoulder roll or ramped position improves airway alignment in obese patients. A lateral position is useful for endoscopy. Position before sedating, not after.

⚡ 5. The "pause and look" rule. After each dose, look at the patient's chest, the waveform, and the face for 60–90 seconds before anyone does anything else.

⚡ 6. Slow ketamine IV push over at least 30–60 seconds. This reduces apnoea and laryngospasm.

⚡ 7. Name the plan out loud. Before starting, the team states: drug, dose, depth aimed for, what we do if the saturation drops below 90%, who calls for help. A 30-second briefing reduces chaos when it matters.

⚡ 8. Use lean body weight for propofol dosing in the obese, and reduce doses in the frail and in those receiving opioids.

⚡ 9. Have one hand on the jaw. In any deeply sedated patient, someone's hands are ready to perform a jaw thrust, not in their pockets.

⚡ 10. Discharge on criteria, not on the clock. Return to baseline cognition, stable vital signs, ability to maintain the airway, pain controlled, adult escort, and written instructions (no driving or important decisions for 24 hours).

---

10. Management Intricacies: Drugs, Sequencing and Pitfalls

10.1 A sequencing template for a painful short procedure (for example shoulder reduction)

1. Assess (AMPLE, airway, STOP-BANG, ASA class). Document baselines.
2. Prepare (SOAPME, capnography on, IV access, reversal agents available, briefing).
3. Consider local or regional options and adjuncts to reduce the sedative requirement.
4. Pre-oxygenate for 3 minutes.
5. Choose the agent:
   • Haemodynamically stable, no contraindication: ketamine is an excellent choice for painful reductions, or propofol at reduced dose with attention to analgesia.
   • Haemodynamically fragile: ketamine at reduced doses or etomidate, with preparation to treat hypotension.
6. Titrate in small aliquots, waiting for each peak.
7. Proceed at the target depth. Maintain verbal contact if aiming for moderate sedation.
8. Recover with continued monitoring until discharge criteria are met.

10.2 Managing common events

Event	First actions	Next steps
Desaturation / apnoea	Stop the procedure and drugs, stimulate, jaw thrust / chin lift, 100% oxygen	Bag-valve-mask with an oral/nasal airway, two-person technique; consider reversal if opioid or benzodiazepine was used
Airway obstruction	Reposition, jaw thrust, suction	Nasopharyngeal or oropharyngeal airway, CPAP via bag-mask; call for airway help
Laryngospasm (high-pitched stridor, or silent obstruction with chest "rocking")	Stop stimulation, 100% oxygen, jaw thrust with firm pressure at Larson's notch, gentle positive pressure	If complete obstruction and hypoxia: deepen with a small dose of propofol; if still refractory, succinylcholine (about 0.1–0.5 mg/kg IV; 4 mg/kg IM if no IV) and secure the airway
Hypotension	Reduce or stop the agent, leg elevation, fluids if appropriate	Vasopressor bolus (phenylephrine or ephedrine); check for other causes
Bradycardia (common with dexmedetomidine, vagal stimulation)	Stop provoking stimulus, ensure oxygenation	Atropine 0.5 mg IV and repeat as needed
Emergence agitation (ketamine)	Quiet environment, reassurance	Small dose of midazolam or propofol if severe
Vomiting	Turn patient lateral, suction	Treat potential aspiration as an emergency
Anaphylaxis	Stop all suspected agents	Intramuscular adrenaline and standard resuscitation

10.3 The pitfalls that recur in case reviews

• No capnography in a deeply sedated patient.
• One person trying to do both the procedure and the sedation.
• Stacking doses before the peak.
• Using sedation to treat pain instead of an analgesic.
• Failing to adjust for age, frailty, OSA or drug interactions.
• No pre-agreed rescue plan or call for help.
• Premature discharge of a patient who was reversed or is still sedated.
• Assuming a patient "woke up quickly" so they are safe. Many serious events occur in recovery.

---

11. When to Escalate and When to Watch

Decisions in sedation turn on trajectory and reserve, not only on a single number.

11.1 Call anaesthesia early (before the procedure or at the first sign of difficulty)

🚩 Escalate before starting if:

• ASA IV or V, or ASA III with major cardiorespiratory instability.
• Anticipated difficult airway, severe OSA, morbid obesity with a difficult airway profile, or inability to lie flat.
• Full stomach with a need for deep sedation and no way to defer, because protecting the airway may require intubation.
• Expected deep sedation or prolonged procedure beyond your competence or monitoring ability.
• Haemodynamic instability requiring vasoactive drugs.
• Late pregnancy, or extremes of age.
• Previous severe sedation reactions or known malignant hyperthermia susceptibility.

🚩 Escalate during the procedure if:

• Hypoxia (SpO₂ below 90%) that does not correct within about a minute of simple manoeuvres (stimulation, jaw thrust, oxygen).
• Absent EtCO₂ waveform with no chest movement despite airway opening.
• Persistent laryngospasm unresponsive to the first-line manoeuvres.
• Persistent hypotension or arrhythmia after simple correction.
• Any need for bag-mask ventilation beyond a brief period, or any thought of intubation.

11.2 When it is reasonable to watch (and why)

Observation	Why it is usually safe to watch	The condition
Brief desaturation that corrects instantly with a jaw thrust or stimulation	Airway tone issue, not a drive failure	You act immediately and observe the trend
Transient mild fall in EtCO₂ amplitude with adequate waveform and chest movement	Often reflects a changing breathing pattern	Confirm trends rather than single values
Mild BP drop (about 10–20% from baseline) in a stable patient	Expected vasodilation with propofol	Patient is asymptomatic and trending upward
Mild ketamine hypertension and tachycardia	Expected sympathomimetic effect	No ischaemic symptoms or severe hypertension
Mild emergence agitation	Self-limiting in most cases	Safe environment and monitoring continue

🎯 The decision principle: Watch only what you can explain and that improves with a single simple action. Escalate anything that needs a second manoeuvre, returns repeatedly, or cannot be explained.

---

12. Summary: The SEDATED Mnemonic and Take-Home Table

12.1 Mnemonic: SEDATED

Letter	Principle	The reminder
S	Stratify	AMPLE, airway (LEMON), STOP-BANG, ASA class, frailty. Who should be sedating?
E	Equipment	SOAPME. Suction, Oxygen, Airway, Pharmacy, Monitors, Equipment.
D	Dose by physiology	Age, frailty, lean weight, interacting drugs. Not the label dose.
A	Analgesia first	Local or regional blocks, opioid or ketamine before the sedative. Pain is not treated by midazolam.
T	Titrate and Time	One drug, one small dose, wait for the peak.
E	End-tidal CO₂	Capnography for every moderate-to-deep sedation. Waveform before saturation.
D	Depth and Discharge	Be ready to rescue one level deeper. Discharge by criteria, with an escort and clear instructions.

12.2 Quick-reference drug card (typical adult starting doses, 70 kg)

Drug	Typical IV dose	Peak	Best for	Main caution
Propofol	0.5–1 mg/kg, then 0.25–0.5 mg/kg	1–2 min	Short non-painful or brief procedures, cardioversion	Hypotension, apnoea, no analgesia
Ketamine	1–1.5 mg/kg slowly (IM 4–5 mg/kg)	1–5 min	Short painful procedures	Emergence, laryngospasm, hypertension
Etomidate	0.1–0.15 mg/kg	1–2 min	Haemodynamic fragility, cardioversion	Myoclonus, adrenal suppression with repeats
Midazolam	0.02–0.03 mg/kg (0.5–1 mg in the old)	3–5 min	Anxiolysis, adjunct	Slow peak, synergy with opioids, accumulation
Fentanyl	0.5–1 mcg/kg (25 mcg increments in the old)	3–5 min	Analgesia	Respiratory depression, rigidity at high dose
Dexmedetomidine	1 mcg/kg over 10 min, then 0.2–1 mcg/kg/h	15–30 min	Cooperative, prolonged sedation	Bradycardia, slow onset
Remimazolam	About 5 mg, then 2.5 mg top-ups (label)	2–4 min	Short procedures, haemodynamic stability	Respiratory depression with opioids

12.3 Ten take-home messages

1. Sedation is a monitoring and rescue discipline before it is a pharmacology one.
2. Capnography detects trouble before the pulse oximeter does.
3. Wait for the peak. Stacking doses is the commonest preventable error.
4. Treat pain with analgesics and blocks, not with deeper sedation.
5. Reduce doses in the old, frail, obese, and those with OSA or interacting drugs.
6. Be ready to rescue one level deeper than your target.
7. Ketamine is a superb painful-procedure agent. Give it slowly and ignore the intracranial-pressure myth.
8. Fasting is a risk-based decision, not a ritual.
9. Reversal agents are a safety net, not a strategy, and they wear off first.
10. Escalate early. Needing a second rescue manoeuvre is the signal to call.

---

References

1. Cravero JP, Blike GT, Beach M, Gallagher SM, Hertzog JH, Havidich JE, et al. Incidence and nature of adverse events during pediatric sedation/anesthesia for procedures outside the operating room: report from the Pediatric Sedation Research Consortium. Pediatrics. 2006;118(3):1087-96.
2. American Society of Anesthesiologists Task Force on Moderate Procedural Sedation and Analgesia. Practice guidelines for moderate procedural sedation and analgesia 2018. Anesthesiology. 2018;128(3):437-79.
3. Green SM, Mason KP, Krauss BS. Pulmonary aspiration during procedural sedation: a comprehensive systematic review. Br J Anaesth. 2017;118(3):344-54.
4. Green SM, Leroy PL, Roback MG, Irwin MG, Andolfatto G, Babl FE, et al. An international multidisciplinary consensus statement on fasting before procedural sedation in adults and children. Anaesthesia. 2020;75(3):374-85.
5. Waugh JB, Epps CA, Khodneva YA. Capnography enhances surveillance of respiratory events during procedural sedation: a meta-analysis. J Clin Anesth. 2011;23(3):189-96.
6. Rex DK, Deenadayalu VP, Eid E, Imperiale TF, Walker JA, Sandhu K, et al. Endoscopist-directed administration of propofol: a worldwide safety experience. Gastroenterology. 2009;137(4):1229-37.
7. Green SM, Roback MG, Kennedy RM, Krauss BS. Clinical practice guideline for emergency department ketamine dissociative sedation: 2011 update. Ann Emerg Med. 2011;57(5):449-61.
8. Green SM, Roback MG, Krauss B, Brown L, McGlone RG, Agrawal D, et al. Predictors of airway and respiratory adverse events with ketamine sedation in the emergency department: an individual-patient data meta-analysis of 8,282 children. Ann Emerg Med. 2009;54(2):158-68.
9. Strayer RJ, Nelson LS. Adverse events associated with ketamine for procedural sedation in adults. Am J Emerg Med. 2008;26(9):985-1028.
10. Sener S, Eken C, Schultz CH, Serinken M, Ozsarac M. Ketamine with and without midazolam for emergency department sedation in adults: a randomized controlled trial. Ann Emerg Med. 2011;57(2):109-14.
11. Miner JR, Moore JC, Austad EJ, Plummer D, Hubbard L, Gray RO. Randomized, double-blinded, clinical trial of propofol, 1:1 propofol/ketamine, and 4:1 propofol/ketamine for deep procedural sedation in the emergency department. Ann Emerg Med. 2015;65(5):479-88.
12. Candiotti KA, Bergese SD, Bokesch PM, Feldman MA, Wisemandle W, Bekker AY; MAC Study Group. Monitored anesthesia care with dexmedetomidine: a prospective, randomized, double-blind, multicenter trial. Anesth Analg. 2010;110(1):47-56.
13. Rex DK, Bhandari R, Desta T, DeMicco MP, Schaeffer C, Etzkorn K, et al. A phase III study evaluating the efficacy and safety of remimazolam (CNS 7056) compared with placebo and midazolam in patients undergoing colonoscopy. Gastrointest Endosc. 2018;88(3):427-37.
14. Pastis NJ, Yarmus LB, Schippers F, Ostroff R, Chen A, Akulian J, et al. Safety and efficacy of remimazolam compared with placebo and midazolam for moderate sedation during bronchoscopy. Chest. 2019;155(1):137-46.
15. Lin Y, Zhang X, Li L, Wei M, Zhao B, Wang X, et al. High-flow nasal cannula oxygen therapy and hypoxia during gastroscopy with propofol sedation: a randomized multicenter clinical trial. Gastrointest Endosc. 2019;90(4):591-601.
16. Godwin SA, Burton JH, Gerardo CJ, Hatten BW, Mace SE, Silvers SM, et al. Clinical policy: procedural sedation and analgesia in the emergency department. Ann Emerg Med. 2014;63(2):247-58.

 

Hypertension: When Not to Treat — and When to Take It Back Down

                                                                    Dr Neeraj Manikath DNB

 

A pragmatic review of therapeutic restraint, measurement discipline, and the craft of deprescribing

 

 

 

Abstract

 

Hypertension remains the world's leading modifiable cause of death, and the reflex to treat every elevated reading is deeply — and rightly — embedded in internal medicine. Yet a substantial minority of blood pressure elevation is artefactual, transient, or low-yield, and a growing population of older adults is harmed not by hypertension but by its treatment. This review examines the two skills that separate competent clinicians from master clinicians in hypertension: knowing when not to start (or intensify) therapy, and knowing when and how to withdraw it. Drawing on recent trials, guideline evolution, and bedside experience, it offers a practical framework — including the LOW PRESSURE mnemonic for deprescribing triggers and the COLD START mnemonic for therapeutic restraint — with decision thresholds, monitoring protocols, and the diagnostic nuances (out-of-office measurement, orthostatic and postprandial physiology, adherence forensics) that make both restraint and withdrawal safe rather than negligent.

 

Key messages

 

1. A blood pressure reading is a screening test, not a diagnosis. Confirm out of office before starting, escalating, or de-escalating.

2. Hypertensive urgency is not an emergency. The fastest way to harm an asymptomatic patient is to slam their pressure down.

3. Deprescribing is a therapeutic experiment with a safety net — one change at a time, monitored, with explicit restart criteria.

4. We treat populations; we care for patients. Absolute risk, frailty, time-to-benefit, and the patient in front of you must override the arithmetic.

 

 

 

1. Introduction: The Patient the Guidelines Forgot

 

Margaret is 84. She arrives in your clinic after an emergency department visit — found on her kitchen floor by a neighbour, transiently confused, CT head unremarkable. Her medication list reads like a cardiology textbook: amlodipine 10 mg, lisinopril 40 mg, metoprolol succinate 95 mg, hydrochlorothiazide 25 mg. Her clinic blood pressure today is 184/88. Her husband died eight months ago; her BMI has drifted from 27 to 21. Her eGFR is 42.

 

The reflex writes itself: uncontrolled hypertension on four agents — this is resistant hypertension; add spironolactone and refer.

 

But you ask one more question: "What does your blood pressure do at home?" Her daughter produces a diary. Mid-morning readings: 96–112 systolic, with lightheadedness before lunch. You walk her to the corridor: lying 150/82, standing 118/68 at two minutes, and she grabs your arm. She has fallen twice this year. Both times, mid-morning.

 

Margaret does not need a fifth drug. Margaret needs subtraction. Stop the thiazide, halve the metoprolol and lisinopril, recheck in two weeks: home systolics 138–148, no dizziness, no falls. Her clinic pressure? 152. And 152 is fine.

 

Now hold the opposite case in mind: the 58-year-old whose 168/98 was called "stress" on three consecutive visits, never confirmed out of office, never worked up, until he woke with a dense hemianopia. Both failures exist, and they share one root: treating or dismissing a number instead of investigating a person.

 

The epidemiology explains why this matters now. Roughly 1.3 billion adults live with hypertension; fewer than a third are controlled. But among adults over 70 taking antihypertensives, symptomatic hypotension, orthostatic intolerance, and injurious falls are endemic — one in three community-dwelling adults over 65 falls each year, and antihypertensives consistently rank among the leading drug classes implicated in emergency department visits for adverse drug events in older adults. The population-level undertreatment of hypertension coexists with individual-level overprescription, particularly at the extremes of age, frailty, and weight.

 

🔑 The thesis of this review: In hypertension, adding a drug is easy medicine. Withholding one — and taking one away at the right moment — is master medicine. Restraint is not neglect when it is reasoned, monitored, and documented; it is neglect only when it is reflexive.

 

This article covers the physiology that makes restraint rational, the seven legitimate reasons not to start, the evidence and protocol for deprescribing, and the diagnostic nuances that keep you honest in both directions.

 

 

 

2. Pathophysiology: Only the Bits That Change Decisions

 

You do not need the full baroreceptor lecture. You need the six physiological facts that change what you do at the bedside.

 

1. The alerting reaction. A clinical setting triggers a sympathetic surge. The result — white coat effect — can inflate systolic pressure by 20–40 mmHg, and the magnitude scales with the office reading itself: the higher the clinic number, the larger the gap between clinic and reality tends to be. Out-of-office pressures correlate with target organ damage and outcomes; isolated office readings correlate with the room. Actionable consequence: every diagnosis, every titration, and every de-escalation decision deserves confirmation by ambulatory (ABPM) or home (HBPM) monitoring.

 

2. Stiff pipes, false numbers. Decades of elastin loss and calcification do two things. First, they can make brachial arteries resist cuff compression — pseudohypertension, the rare syndrome where the cuff reports 180 and the artery carries 140. Suspect it in the elderly with impressive clinic pressures but no target organ damage, or hypotensive symptoms on "appropriate" therapy. Second, stiffness produces isolated systolic hypertension with a wide pulse pressure: systolic 172, diastolic 58. Chasing the systolic drags the diastolic toward the floor — and the left ventricle perfuses in diastole.

 

3. The J-curve is real enough to respect. The HOT trial found the lowest cardiovascular event rates at a diastolic around 82–83, with risk rising below ~70, especially in coronary disease — coronary perfusion pressure is diastolic pressure minus LV end-diastolic pressure. Whether the J-curve is causation or reverse causation (sick patients run low pressures) is still debated; the bedside consequence is not: in known coronary disease, think twice before driving diastolic pressure below 60–65.

 

4. Autoregulation shifts rightward. Chronic hypertension shifts cerebral and renal autoregulatory curves upward. The chronic hypertensive brain perfused adequately at a mean pressure of 120 may drop below its autoregulatory floor if you halve that pressure in an hour. This single fact underwrites the ≤25% mean arterial pressure reduction in the first hour rule for emergencies, the condemnation of sublingual nifedipine, and permissive hypertension in acute stroke.

 

5. The baroreflex ages. Blunted baroreflex buffering means the elderly hypertensive patient swings: supine 180, standing 110, post-lunch 100. The same reflex that kept her upright at 60 now under-corrects at 85. In patients over ~75, the standing and postprandial numbers are as decision-relevant as the sitting one.

 

6. Late-life decline can be prodrome, not success. Longitudinal cohorts (including ARIC) show that blood pressure trajectories that fall in late life, particularly from a midlife hypertensive baseline, are associated with increased dementia risk — plausibly early neurodegeneration of autonomic centres. A drifting-down systolic in an 86-year-old is not a treatment triumph to be consolidated; it may be a disease signal to be respected. Conversely, the same populations show midlife hypertension drives late-life dementia — treat the 55-year-old's 150 rigorously; be humble about the 85-year-old's.

 

🧠 One-line physiology recap: treat the perfusion, not the arithmetic. The cuff measures pressure; the brain, heart, and kidneys consume flow.

 

 

 

3. When NOT to Start (or Intensify): Seven Legitimate Reasons to Hold Fire

 

3.1 The number isn't real

 

Before any first prescription, eliminate measurement error — the cheapest "diagnosis" you will ever make:

 

● Technique: 5 minutes seated, back supported, feet flat, arm supported at heart level, correct cuff (bladder encircling ~80% of arm circumference), no talking, no caffeine for 30 minutes, bladder empty, two readings a minute apart. Each violation alone shifts readings by 5–15 mmHg — a small cuff on a large arm can overestimate systolic pressure by 10–15 mmHg, manufacturing hypertension out of adipose tissue.

● White coat hypertension (elevated office, normal out-of-office) accounts for perhaps 15–30% of clinic diagnoses. Meta-analysis shows it is not risk-free — event risk sits roughly 20–40% above true normotension, driven by the fact that white coat cohorts often carry weight, glucose, and age [7]. But in the patient without target organ damage and with low absolute risk, drug therapy buys side effects without meaningful absolute benefit. The correct prescription is lifestyle plus surveillance — annual ABPM or HBPM — upgrading to pharmacotherapy if risk factors or organ damage accumulate.

● Pseudohypertension (Osler's manoeuvre still positive — a palpable radial when the cuff is pumped above systolic) in the severely arteriosclerotic elderly.

 

⚠️ Guardrail: restraint must not curdle into neglect. If the office numbers are normal but the patient is not — unexpected left ventricular hypertrophy, microalbuminuria, retinopathy, a lacunar stroke — suspect masked hypertension (normal office, elevated out-of-office), whose risk approaches sustained hypertension. Confirm with ABPM before accepting reassurance.

 

3.2 The number is real but transient

 

Catecholamines are honest reporters of misery. Acute pain, anxiety and panic attacks, urinary retention (check the elderly man before you check the cuff), a sleepless hospital night, hypoglycaemia, alcohol withdrawal, cocaine or amphetamines, acute post-operative states, thyroid storm — all elevate pressure honestly. Hospital pressures run 5–10 mmHg above home baselines on pain and sleep deprivation alone. Treat the cause; treat the pain; catheterise the bladder; sedate the withdrawal — then remeasure. A new antihypertensive started for a transient surge becomes a lifelong prescription at the next medication reconciliation, written by nobody, defended by no one, stopped by nothing.

 

Drug-induced elevation deserves its own scan before any new prescription: NSAIDs, systemic corticosteroids, oral contraceptives, pseudoephedrine and nasal decongestants, stimulants for ADHD, venlafaxine and duloxetine, erythropoietin, cyclosporine and tacrolimus, anabolic steroids, high-dose caffeine, energy drinks, licorice. The first prescription is withdrawal of the culprit, not addition of an antidote.

 

3.3 Real and sustained — but the risk doesn't yet justify a pill

 

Here the guidelines genuinely disagree, and you should know why. The 2023 ESC guideline defines a category of elevated blood pressure (120–139/70–89) for which drugs are not recommended, and for grade 1 hypertension (140–159/70–99) without established disease, high calculated risk, or hypertension-mediated organ damage, recommends a lifestyle-first window of some months before reconsidering drugs [5]. The 2017 ACC/AHA guideline labels the same territory "stage 1" and recommends drugs only with a 10-year ASCVD risk ≥10%, diabetes, CKD, or organ damage; below that, lifestyle with reassessment at 3–6 months [6]. The 2023 ESH guideline similarly anchors initiation to grade, risk, and organ damage [4].

 

The trial truth beneath the divergence: a Cochrane review of older trials found no demonstrable outcome benefit from treating low-risk mild hypertension, while the Blood Pressure Lowering Treatment Trialists' Collaboration meta-analysis showed a roughly constant ~10% relative reduction in major events per 5 mmHg systolic reduction — but relative benefit is not absolute benefit, and in a patient whose 10-year risk is 2%, a 10% relative reduction is a rounding error compared with a real risk of symptomatic hypotension, electrolyte disturbance, or falls [3].

 

The synthesis: thresholds are population heuristics. At the bedside, compute absolute risk, hunt for organ damage (see §9), and make a shared decision. And set a date — "we re-check in 3 months with a home diary" — so restraint has a review date and cannot quietly become abandonment.

 

3.4 Urgency is not an emergency

 

Hypertensive urgency (≥180/120 without acute target organ damage) is not treated with intravenous drugs. There is no endpoint that improves when you slam the pressure of an asymptomatic person; historical series are full of iatrogenic strokes and myocardial infarctions from rapid reduction, and sublingual nifedipine — the drug that drops like a stone — has been condemned for decades.

 

The urgent patient's real problem is usually one of three things: they ran out of their medications (a dose of their own drug solves it), they are in pain or distress (treat that), or they are lost to follow-up (that is the disease). The correct disposition: restart or titrate oral therapy, reassess within the hour, and secure a clinician contact within 24–72 hours — ideally with a same-week primary care visit. The treatment of urgency is a follow-up plan.

 

3.5 Permissive hypertension and contexts where speed kills

 

● Acute ischaemic stroke, no reperfusion therapy: do not treat below 220/120 systolic. The elevated pressure is perfusing the penumbra.

● After thrombolysis: lower to <185/110 before lysis and keep <180/105 for 24 hours — no lower than needed.

● Intracerebral haemorrhage: controlled acute reduction to roughly 130–140 systolic is safe and guideline-supported; overshooting below ~130 adds renal and ischaemic harm without benefit.

● Aortic dissection — the great exception: systolic <120 within 20 minutes, beta-blockade first (unopposed vasodilation accelerates the tear).

● Severe pre-eclampsia/eclampsia — the second exception: ≥160/110 should be lowered promptly with labetalol, hydralazine, or nicardipine, plus magnesium. (Pregnancy is the one realm where "urgent" numbers are emergencies.)

● Autonomic failure: the patient with supine 190 and standing 95. Treat the standing number; tolerate the supine one.

● Perioperative: do not initiate beta-blockade de novo to "protect" non-cardiac surgery — POISE showed excess mortality and stroke, mediated by hypotension and bradycardia [15]. Continue established therapy, but note that newer factorial evidence (POISE-3) supports holding ACE inhibitors/ARBs on the morning of surgery to avoid intraoperative hypotension.

● Fixed obstruction (severe aortic stenosis, hypertrophic cardiomyopathy): preload-dependent patients tolerate aggressive vasodilation poorly; lower gently, if at all, in the acute setting.

 

3.6 The best first prescription may not be a pill

 

Three conditions routinely masquerade as "more drugs needed":

 

1. Obstructive sleep apnoea — present in the majority of resistant hypertension. Screen (STOP-Bang ≥3, or any snoring hypertensive with a body mass index >30). CPAP lowers pressure only ~2–3 mmHg on average — set expectations — but it is part of the causal treatment.

2. Primary aldosteronism — ~1 in 10 hypertensives overall, up to 1 in 5 among resistant cases. Hypertension with spontaneous or easily provoked hypokalaemia, resistant hypertension, or an adrenal incidentaloma deserves an aldosterone-renin ratio, prepared correctly (see §9).

3. Atherosclerotic renal artery stenosis — stenting added nothing to modern medical therapy in CORAL [14]. Reserve revascularisation for refractory hypertension despite optimal drugs, recurrent flash pulmonary oedema, rapidly declining renal function, or bilateral/solitary-kidney stenosis.

 

3.7 When time-to-benefit runs out

 

The benefits of blood pressure control accrue over years. In advanced frailty (Clinical Frailty Scale ≥7), involuntary weight loss, life expectancy measured in months, or advanced dementia, the calculus collapses: the falls and hypotension arrive this quarter; the stroke reduction was never coming. The STOPP criteria and Choosing Wisely campaigns converge here. A systolic of 165 in a comfortable 92-year-old with advanced dementia is not a task. It is a number.

 

🧊 If you remember nothing else — a COLD START is when you don't start:
C — Cuff/measurement artefact (miscuffing, wrist misuse, Osler's sign)
O — Occasional elevations only (white coat, pain, panic, full bladder, hospital nights)
L — Low absolute risk, early stage, no organ damage: lifestyle-first window
D — Drug-induced or removable cause (NSAIDs, steroids, stimulants, OSA, aldosteronism, stenosis)
S — Stroke and other permissive acute contexts — don't slam the pressure
T — Terminal frailty or goals of care that outrank long-horizon benefit
A — Autonomic failure (supine-high, standing-low — treat the standing)
R — Rapid lowering is the real risk (urgency ≠ emergency)
T — Thresholds not met: no organ damage, no diabetes/CKD, risk <10%

 

 

 

4. When to Decrease or Stop: Deprescribing as High-Skill Medicine

 

4.1 The asymmetry of our habits

 

Every guideline teaches initiation; almost none taught withdrawal until very recently. We celebrate "intensification" and quietly fear that reducing therapy is abdication. Reframe it: deprescribing is a therapeutic trial of less medicine, run with the same rigour as starting one — defined trigger, single intervention, monitoring interval, and a restart plan.

 

4.2 The triggers

 

Group them into five families:

 

● Symptoms: orthostatic dizziness, postprandial lightheadedness, fatigue, syncope, falls, symptomatic bradycardia, drug-specific adverse effects.

● Numbers: sustained home/ambulatory systolic <110, or <130 with hypotensive symptoms; bradycardia; over-treatment demonstrated on ABPM (including nocturnal hypotension).

● People: age ≥80, frailty, involuntary weight loss, nursing home residence, polypharmacy.

● Events: any fall, any hospitalisation, stroke, gastrointestinal bleed, major surgery — each is a recalibration moment. Post-hospitalisation, chronic regimens routinely need reduction.

● Resolved causes: effective CPAP, adrenalectomy for unilateral aldosteronism, NSAID cessation, significant alcohol reduction, major weight loss or bariatric surgery, correction of thyroid disease. After bariatric surgery especially, be proactive — the weight loss will deprescribe for you, in the form of syncope, if you don't get there first.

 

4.3 What the trials actually show

 

Three randomised trials have now tested the heresy:

 

● OPTIMISE [9]: community-dwelling adults ≥80 (or ≥65 with frailty) already controlled (systolic <140) on treatment. A structured reduction strategy maintained control at 12 weeks — systolic pressure was, if anything, marginally lower than usual care — with no excess of serious harms; a minority required reinstatement.

● SIMPLIFY [10]: simplifying the regimen (dropping one agent or consolidating into a combination pill) maintained blood pressure control at 12 months with fewer daily tablets.

● EMERGE [11]: nursing home residents; a structured deprescribing intervention increased successful dose reduction and cessation with maintained control and no short-term excess harm.

 

Read these honestly: short follow-up, selected populations, blood pressure — not cardiovascular events — as the endpoint. What they license is monitored deprescribing in appropriate candidates; what they do not license is casual cessation. The discipline is in the protocol.

 

4.4 The protocol (do it exactly this way)

 

1. Establish the trigger with out-of-office data: a 7-day home diary plus lying/standing vitals (and a post-lunch reading in the symptomatic elderly).

2. Choose one drug. Rank by (weakest indication + loudest adverse effects): alpha-blockers and centrally acting agents first; beta-blockers next if there is no compelling indication (no ischaemia, no heart failure, no rate requirement); then dose reduction of the remainder. The last to go — or never casually — are agents protecting a specific organ: ACE inhibitor/ARB in proteinuric CKD or heart failure, beta-blocker after infarction, mineralocorticoid antagonist in heart failure.

3. One change at a time. Reduce the dose 25–50% or stop a single agent.

4. Taper the two dangerous classes. Beta-blockers: halve the dose every 5–7 days over 1–2 weeks — abrupt withdrawal causes rebound ischaemia and tachyarrhythmia. Clonidine: taper, or reinstate via patch — abrupt cessation causes rebound hypertension with tachycardia and agitation within 24–72 hours. Diuretics, calcium channel blockers, and RAS blockers can be stopped without rebound (watch potassium and volume status).

5. Monitor and define restart criteria. Home readings twice daily for 1–2 weeks, then weekly to 6–8 weeks, with a written contract: "Restart / call us if home systolic stays above 155 for three days, or if headaches/visual symptoms return." In higher-risk patients or those with prior organ damage, use a lower trigger (~140–150).

6. Document the reasoning and hand it forward — discharge summaries, pharmacy, the next clinician. An undocumented deprescription is indistinguishable from an error by morning.

 

🪙 Grand-round rule: The best time to deprescribe is before the fracture — not after the hip has done your outcome analysis for you.

 

4.5 Event-driven windows you will meet on the wards

 

● One to three months after stroke: blood pressure falls spontaneously after the subacute phase; many patients are overtreated by the three-month review. Reassess deliberately.

● After any hospitalisation in an older adult: the "post-hospital syndrome" — volume shifts, weight loss, deconditioning — recalibrates tolerance. A discharge is a deprescribing opportunity; institutionalise a medication review before the patient leaves.

● After intercurrent illness with weight loss, dehydration, or acute kidney injury: reduce, then rebuild if needed.

● Palliative and end-of-life care: as goals shift, the antihypertensive is often among the first medications to retire gracefully. Stopping it there is not defeat; it is precision.

 

 

 

5. Clinical Pearls 🪙

 

1. The higher the office reading with a normal home diary, the bigger the white coat effect. Treat the trend, never the peak.

2. A "normal" pressure in a chronically hypertensive patient can be relative hypotension. Always ask: "What does it usually run at home?" Baseline is context.

3. Hospital hypertension is usually pain, a full bladder, or a sleepless night. Treat those, remeasure, and expect a 5–10 mmHg ward-to-home drift before you touch the chart.

4. Falls on antihypertensives get labelled TIAs. The medication list is cheaper than the stroke workup — and more often guilty. Ask about the timing of falls relative to the morning dose.

5. In acute ischaemic stroke, the high pressure is often the penumbra's lifeline. Treat the reperfusion plan, not the monitor.

6. A creatinine rise ≤30% after starting an ACE inhibitor/ARB is physiology, not nephrotoxicity — and predicts long-term renal benefit. Beyond 30%, or potassium >5.6: think bilateral renal artery stenosis, volume depletion, NSAIDs.

7. Postprandial dizziness in an elderly patient is pressure, not "low sugar." Measure before and one hour after lunch.

8. Beta-blocker or clonidine withdrawal mimics phaeochromocytoma — hypertensive, tachycardic, diaphoretic agitation. Check the med list before ordering metanephrines.

9. Two tablets taken religiously beat five taken intermittently. Non-adherence masquerades as resistance and escalates prescriptions (see Oyster #1).

10. In the very old, a systolic of 150 that keeps someone upright outperforms a 125 that puts them on the floor.

11. If the cuff disagrees with the patient, believe the patient. Symptoms with "normal" readings mean you measured the wrong posture, the wrong time, or the wrong arm.

12. White coat hypertension is a risk licence, not a prescription pad — lifestyle, surveillance, and vigilance for the risk that accrues beneath it [7].

 

 

 

6. Oysters 🦪 — The Gems Most Clinicians Miss

 

1. Urine toxicology as adherence forensics. Liquid chromatography assays detect absent parent drugs in up to half of "resistant" referrals [8]. Before you add spironolactone, consider that the fourth drug may be un-swallowed.

2. Licorice (glycyrrhizin) — apparent mineralocorticoid excess. Hypertension, hypokalaemia, metabolic alkalosis, suppressed renin and aldosterone. Ask about herbal teas, confectionery, and certain chewing tobaccos in every odd refractory case.

3. The missed clonidine patch. Admission to hospital without the patch = rebound hypertension + tachycardia + agitation at 24–72 hours, routinely misread as sepsis or alcohol withdrawal. Reinstate the patch.

4. The auscultatory gap. In stiff elderly arteries, Korotkoff sounds vanish then reappear; auscultation underestimates systolic — the patient looks "controlled" while running high. Palpate first, or use oscillometry.

5. Late-life decline as dementia prodrome. A falling systolic trajectory in an octogenarian predicts cognitive decline [12]; pharmacologically "correcting" it treats neither heart nor brain. Conversely, a stable 150 in an 88-year-old with intact cognition is not an emergency.

6. Chronotherapy was a mirage. The TIME trial found no benefit to bedtime dosing [13]; the spectacular Hygia results championing it have been retracted. Spend the counselling minutes on adherence, not clocks.

7. Adrenal incidentaloma + hypertension = an aldosterone-renin ratio, not reassurance. Primary aldosteronism is common and eminently treatable — and unilateral disease cured surgically is a deprescription bonanza.

8. Alpha before beta in catecholamine excess. Never block beta first in suspected phaeochromocytoma — unopposed alpha activity precipitates crisis. (The mirror image: beta before vasodilator in aortic dissection.)

9. Forearm and ankle cuffs overestimate systolic; wrist monitors held below heart level inflate readings — a quiet factory of phantom "uncontrolled hypertension" in obese and arrhythmic patients.

10. Acarbose for postprandial hypotension. By delaying glucose absorption it blunts the splanchnic vasodilatory surge — a diabetes drug repurposed for the syncopal luncher, alongside small frequent low-carbohydrate meals and coffee with meals.

11. Post-bariatric patients crash on their old regimens. Put deprescribing on the surgical follow-up checklist before the syncope writes it for you.

12. The silent initiators: venlafaxine, erythropoietin, cyclosporine, anabolic steroids, energy drinks, yohimbine, and high-dose caffeine. The "resistant" hypertensive's review of systems should include the gym bag and the supplement shelf.

 

 

 

7. Clinical Hacks & Tips ⚡

 

1. The three magic questions, every hypertensive visit: (1) Do you get dizzy when you stand or mid-morning? (2) Have you fallen in the past year? (3) What does your pressure read at home when you feel unwell? Three questions, most overtreatment detected.

2. Never titrate on a single office reading — in either direction. Escalation and de-escalation both require a 7-day home diary or ABPM. One reading is a screening test.

3. The 5-minute discipline: rest 5 minutes, back supported, feet flat, arm at heart level, correct cuff, no talking, bladder empty, two readings. Discard day 1 of any home diary (anxiety inflation).

4. Check both arms once; follow the higher arm forever. A persistent ≥10–15 mmHg inter-arm difference is itself a vascular risk marker.

5. Pharmacy refill math before the fourth drug: proportion of days covered <80% = non-adherence until proven otherwise. Frame it without shame: "These drugs are only difficult because of how they're taken — let's fix the regimen, not blame you."

6. Dose-timing choreography: don't stack every vasodilator at 7 a.m. (the 10 a.m. swoon is pharmacology, not mystery); morning diuretics to protect sleep; hold the pre-lunch diuretic in postprandial hypotension. Timing for symptom smoothing — not for outcome magic [13].

7. Subtract the weakest indication and loudest side effect first; halve doses before stopping long-acting agents.

8. The restart contract, in writing: explicit thresholds and a phone number. Deprescribing without a safety net is a gamble; with one, it's a trial.

9. Urgency hack: a dose of the patient's own missed medication, a calm room, analgesia if hurting, a toilet visit, and a 72-hour appointment outperforms any intravenous bolus ever made.

10. Write the target and the floor on the chart: "Systolic 125–135 at home; hold amlodipine if systolic <100; call if <95 or dizzy." Night teams need the ceiling and the cellar.

11. The brown-bag visit: every box, bottle, tea, and "energy capsule" on the table — fifteen minutes that has solved more "resistant hypertension" than any fourth agent.

12. Pair the examination with the vitals in patients over 75: standing pressure plus a get-up-and-go test. If both are abnormal, the next prescription is subtraction, not escalation.

 

 

 

8. State of the Art: What Is Actually Changing Practice

 

The measurement paradox of the intensive-target era. SPRINT's dramatic benefit at a systolic target of 120 was achieved with unattended automated office measurement [1] — a modality that runs lower than routine clinic measurement. A "SPRINT 120" approximates 130–140 by conventional office standards. Before you chase 120 on a manual cuff, know which instrument produced your target. STEP then extended intensive control benefit to patients aged 60–80 already on treatment [2] — but its robust elders are not your frail faller, and neither trial enrolled the standing-hypotensive nursing home resident. The intensive-target evidence is real; its generalisability boundary is equally real.

 

The guideline schism, translated for the bedside. ESC 2023 created the "elevated blood pressure" category and withheld drugs for it; ACC/AHA 2017 calls the same numbers stage 1 and treats selectively [4,5,6]. The reconciliation is not to pick a tribe but to compute: relative risk reduction is stable per 5 mmHg [3]; absolute benefit decides the individual. Intensity where absolute risk is high; restraint where it is negligible — and always with a review date.

 

Deprescribing has entered the randomised-trial era. OPTIMISE, SIMPLIFY, and EMERGE [9,10,11] transformed withdrawal from folklore into a monitored, evidence-informed procedure (see §4.3). The next iteration of major guidelines is expected to formalise it; be early, be careful.

 

Chronotherapy has left the building. TIME: no benefit to evening dosing [13]; Hygia: retracted. Case closed.

 

Perioperative philosophy has inverted. From POISE's "don't start beta-blockers" [15] to POISE-3's hypotension-avoidance signals, the field has learned that intraoperative hypotension is the enemy — hold the morning ACE inhibitor/ARB, protect the pressure that perfuses.

 

New molecules for true resistant hypertension — after the homework. Aprocitentan (dual endothelin antagonist, ~5 mmHg placebo-corrected in PRECEDENT), aldosterone-synthase inhibitors in development, and renal denervation with modest, guideline-circumscribed effects. All arrive after adherence forensics, sleep studies, aldosterone ratios, and spironolactone (which beat beta-blockade and alpha-blockade head-to-head as the fourth drug in PATHWAY-2). Devices and novel molecules are the last mile, not the shortcut.

 

The brain has entered the target-organ conversation. Midlife hypertension drives late-life dementia; SPRINT-MIND signalled cognitive benefit from intensive control; and late-life hypotension shadows neurodegeneration [12]. The clinical distillation: treat hypertension like your middle-aged patient's future brain depends on it — because it does — and respect the falling pressure of the old as information, not obligation.

 

 

 

9. Diagnostic Nuances: Where Good and Great Separate

 

The consultation is the instrument. Use devices on validated lists (e.g., the US Validated Device Listing, STRIDE BP), not whatever the pharmacy discounted. Cuff size above pride. In atrial fibrillation, average several oscillometric readings — single beats mislead.

 

Out-of-office thresholds, on the tip of your tongue: ABPM — 24-hour ≥130/80, awake ≥135/85, asleep ≥120/70; HBPM — ≥135/85. Request ABPM specifically when you suspect white coat or masked patterns, episodic elevation (consider phaeochromocytoma), nocturnal dipping questions, or hypotensive symptoms on treatment — the last is the most underused indication in medicine. A reverse dipper (pressure rising at night) carries the worst prognosis of the dipping patterns.

 

Postures and meals are diagnostic manoeuvres. Orthostatic vitals: lying, then standing at 1 and 3 minutes; a delayed drop may need a 10-minute stand. Postprandial: measure before and one hour after lunch — the meal is the stress test.

 

Look at the patient, not just the number. Radio-femoral delay or leg-blood-pressure discrepancy in the young hypertensive (coarctation until excluded); abdominal or flank bruit with recurrent flash pulmonary oedema (renal artery stenosis); purple striae and proximal myopathy (Cushing); café-au-lait patches and neurofibromas (NF-1 and phaeochromocytoma risk); thyromegaly with a stare.

 

Target organ workup that changes decisions: urine albumin-creatinine ratio (≥3 mg/mmol territory escalates risk and RAS-blocker indications), ECG with Cornell/Sokolow-Lyon criteria, echocardiography when symptoms or murmur demand it, eGFR trajectory (a slope, not a point), and fundoscopy reserved for the emergency department evaluation, not routine clinics.

 

The aldosterone-renin ratio, prepared properly. Correct potassium first; interpret with drug knowledge — beta-blockers and clonidine suppress renin (false positives), ACE inhibitors/ARBs raise it (false negatives), diuretics and mineralocorticoid antagonists confound; verapamil ± hydralazine is the classic "clean" bridge. Screen before escalating the resistant patient, not after the fourth drug.

 

Metanephrines, timed sensibly: not during the acute catecholamine storm; after stabilisation, off interfering agents (caffeine, paracetamol, some antidepressants).

 

Adherence forensics as a diagnostic modality: refill records, pill counts, witnessed dosing, and — where available — urine drug assays [8]. "Resistant hypertension" that is unswallowed is the most treatable form in the catalogue.

 

 

 

10. Management Intricacies: Getting It Right When You Do Treat

 

Choose by compelling indication, not by habit. Post-infarction and heart failure choose the beta-blocker and RAS blockade; proteinuric CKD chooses the ACE inhibitor/ARB; the elderly with isolated systolic hypertension favours a calcium channel blocker plus a thiazide-like diuretic (indapamide or chlorthalidone over hydrochlorothiazide); pregnancy chooses labetalol, nifedipine, or methyldopa and forbids RAS blockade. Without a compelling indication, beta-blockers are not first-line, and alpha-blockers never were (ALLHAT settled that).

 

Combine early, dose gently. A low-dose two-drug single pill achieves control faster with fewer side effects than heroic sequential monotherapy titration — and single-pill combinations are quietly an adherence intervention (SIMPLIFY logic [10]).

 

Switch, don't stack, at class ceilings. Amlodipine-related ankle oedema is dose-dependent and mitigated by a RAS blocker — halve the amlodipine or add the ACE inhibitor rather than pushing 10 mg and adding furosemide for iatrogenic oedema. ACE-inhibitor cough resolves by substitution with an ARB, not by addition of a cough suppressant or a third agent. Verapamil constipation, thiazide gout, spironolactone breast tenderness: each is a signal to redesign the regimen, not to anaesthetise its complication.

 

Timing and the interactions that ambush. Once-daily morning dosing for most; morning diuretics; evening redistribution only to smooth symptomatic troughs [13]. Remember the ambush list: PDE-5 inhibitors + nitrates (life-threatening hypotension — a one-week separation is not enough; the combination is contraindicated), amlodipine + simvastatin (20 mg cap), thiazide or ACE inhibitor + lithium (toxicity), NSAIDs + everything (blunt the whole regimen, raise pressure 3–5 mmHg), atenolol in progressive CKD (renal clearance, accumulation, bradycardia).

 

Emergency pharmacology in one breath. Titratable infusions — nicardipine or labetalol (esmolol where heart rate must be controlled first, clevidipine where available). Reduce mean arterial pressure ≤25% in the first hour, toward ~160/100 over 2–6 hours, normalised gently over 24–48 hours. Avoid: sublingual nifedipine (precipitous, unpredictable), reflex hydralazine boluses in coronary patients, enalaprilat (slow and erratic), nitroprusside at high dose or prolonged infusion (cyanide, raised intracranial pressure, and avoid in pregnancy).

 

The diastolic floor in wide pulse pressure. When systolic 175 meets diastolic 55 in a patient with coronary disease, the target is a range negotiated with the diastolic — accept a higher systolic, or select agents (thiazide-like diuretic, calcium channel blocker, RAS blocker) and avoid excess beta-blockade. Write the floor on the chart: "do not drive diastolic below 60."

 

 

 

11. When to Escalate vs When to Watch: Thresholds with Reasoning

 

Escalate within minutes to hours — the true emergencies, where speed is life: hypertensive encephalopathy; intracerebral haemorrhage; aortic dissection (systolic <120 within 20 minutes, beta first); acute coronary syndrome with pressure-driven ischaemia; acute pulmonary oedema; eclampsia/severe pre-eclampsia (≥160/110 promptly, magnesium aboard); malignant-phase hypertension with papilloedema or rapid renal decline. Monitored setting, titratable infusion, ≤25% mean arterial reduction — the dissection and pre-eclampsia exceptions above.

 

Escalate over weeks — confirmed sustained elevation above an individualised target, on out-of-office measurement, tolerated by the patient, with rising absolute risk or new organ damage. Then do it properly: single-pill combinations, real doses, real intervals, and review of adherence each step.

 

The resistant-hypertension ladder, in order: (1) confirm with ABPM and exclude pseudoresistance (white coat, non-adherence, measurement error); (2) remove offenders, treat OSA, screen aldosterone; (3) optimise a true triple regimen including a diuretic; (4) spironolactone 25–50 mg with potassium/creatinine checks at 1, 4, and 8–12 weeks (PATHWAY-2's evidence); (5) only then aprocitentan, devices, or specialist referral.

 

Watch, with a diary and a date: low-risk early hypertension on lifestyle (reassess 3–6 months); white coat hypertension with low risk (annual out-of-office surveillance); urgency without organ damage (contact within 24–72 hours); the subacute stroke patient (reassess at 1–3 months — the pressure often falls by itself); the treated-at-target patient whose only "uncontrolled" numbers live in your clinic (home diary, no titration).

 

Scenario	The reflex	The master's move	Threshold / reasoning
Clinic 185/110, asymptomatic, no organ damage	IV bolus, admit	Confirm; oral therapy/restart own meds; follow-up 24–72 h	Urgency ≠ emergency; fastest harm is iatrogenic
High office, normal home diary	Titrate	White coat: lifestyle + surveillance	Annual ABPM; treat if risk/organ damage accrues
Normal office, but LVH/albuminuria/stroke	Reassure	ABPM — masked hypertension	Masked risk ≈ sustained hypertension
Dizzy elder on 4 drugs, home systolic 96–110	Add a 5th (clinic reads 184)	Deprescribe one; monitor	Systolic <110 or symptoms → subtract
Acute ischaemic stroke, 205/105, no lysis	Lower to 140	Permissive hypertension	Treat only >220/120; post-lysis <185/110 → <180/105
Aortic dissection	Gradual reduction	Fast: systolic <120 in 20 min, beta-blocker first	Wall stress is the killer
Creatinine +25% on new ACE inhibitor	Stop the ACE inhibitor	Continue; monitor	≤30% rise is expected physiology
"Resistant" on 3 drugs	Add a 4th	Verify adherence, OSA, aldosterone, ABPM first	Up to half of resistance is unswallowed [8]
Frail 88-year-old, systolic 155, no symptoms	Intensify	Watch; deprescribing review	Time-to-benefit absent; falls risk imminent
6 weeks post-stroke, systolic now 128 on 3 drugs	Continue all	Reassess and reduce	Spontaneous post-stroke decline
Post-bariatric surgery, dizzy	Continue regimen	Proactive deprescribing	Weight loss lowers pressure dramatically
Hospital night, 170s + pain	Add a drug	Analgesia, sleep, toilet, remeasure	Iatrogenic catecholamines

 

 

 

 

12. The Mnemonic: LOW PRESSURE

 

The triggers to reduce or withdraw antihypertensive therapy:

 

● L — Lightheadedness on standing (or after meals): orthostatic and postprandial hypotension

● O — Over-treatment symptoms: fatigue, syncope, "off days," falls

● W — Walking impaired: frailty, weight loss, nursing home residence

● P — Pressure low at home: sustained systolic <110 (or <130 with symptoms)

● R — Reversible cause removed: OSA treated, adrenalectomy, NSAIDs stopped, alcohol cut, major weight loss

● E — Event just occurred: stroke, GI bleed, hospitalisation, surgery — recalibrate everything

● S — Systolic sliding downward in late life (prodrome, not triumph — don't chase it)

● S — Side effects speaking: cough, oedema, bradycardia, erectile dysfunction

● U — Unnecessary polypharmacy: no compelling indication for each agent

● R — Rebound-risk drugs handled properly: taper beta-blockers and clonidine, never stop abruptly

● E — End of life or advanced dementia: comfort is the indication now

 

Pair it with COLD START (§3) — the circumstances in which starting is the wrong reflex — and you carry both halves of this discipline in your pocket.

 

 

 

13. Conclusion: The Courage to Subtract

 

Hypertension medicine celebrates the courage to treat. This review has argued for its twin virtues, less photogenic and more difficult: the courage to withhold when the number is artefact, transient, low-yield, or outranked by frailty and time; and the humility to subtract when the treatment that once protected becomes the thing that fells.

 

Both virtues run on the same fuel — measurement you can trust, physiology you respect, risk you actually compute, and follow-up you actually book. Restraint without monitoring is neglect; deprescribing without a safety net is a gamble. But done properly, holding fire and easing off are not lesser forms of care. They are the marks of the clinician who has understood, at last, that the point was never the number on the cuff.

 

It was always the person attached to the arm.

 

 

 

References

 

1. Wright JT Jr, Williamson JD, Whelton PK, et al.; SPRINT Research Group. A randomized trial of intensive versus standard blood-pressure control. N Engl J Med. 2015;373(22):2103–16.

2. Zhang W, Zhang S, Deng Y, et al.; STEP Study Group. Trial of intensive blood-pressure control in older patients with hypertension. N Engl J Med. 2021;385(14):1268–79.

3. Blood Pressure Lowering Treatment Trialists' Collaboration. Pharmacological blood pressure lowering for primary and secondary prevention of cardiovascular disease across different levels of blood pressure: a meta-analysis of randomised trials. Lancet. 2021;397(10282):1625–36.

4. Mancia G, Kreutz R, Brunström M, et al. 2023 European Society of Hypertension guidelines for the management of arterial hypertension. J Hypertens. 2023;41(12):1874–2071.

5. Williams B, Mancia G, Spiering W, et al. 2023 ESC guidelines for the management of elevated blood pressure and hypertension. Eur Heart J. 2023;44(38):3826–932.

6. Whelton PK, Carey RM, Aronow WS, et al. 2017 ACC/AHA guideline for the prevention, detection, evaluation, and management of high blood pressure in adults. Hypertension. 2018;71(6):e13–e115.

7. Cohen JB, Lotito ML, Mulvaney C, et al. Cardiovascular events and mortality in white coat hypertension: a systematic review and meta-analysis. Ann Intern Med. 2019;171(3):169–79.

8. Jung O, Gechter JL, Wunder C, et al. Resistant hypertension? Assessment of adherence by toxicological urine analysis. J Hypertens. 2013;31(4):766–74.

9. Sheppard JP, Burt J, Lown M, et al. Effect of antihypertensive medication reduction versus usual care on blood pressure control in patients aged 80 years and older: the OPTIMISE randomised controlled trial. BMJ. 2020;370:m1867.

10. Ahmed A, Zaman M, Patel P, et al. Effect of antihypertensive medication simplification versus continuation of the usual regimen on systolic blood pressure: the SIMPLIFY randomised clinical trial. JAMA Netw Open. 2022;5(7):e2220390.

11. Eamer EM, Al-Jubouri H, Mayrhofer A, et al. Effect of a multicomponent antihypertensive deprescribing intervention in nursing home residents: the EMERGE randomised clinical trial. JAMA Intern Med. 2024;184(6):628–36.

12. Walker KA, Sharrett AR, Wu A, et al. Association of midlife to late-life blood pressure patterns with incident dementia. JAMA. 2019;322(6):535–45.

13. TIME Study Group. Administration of blood pressure-lowering medications in the morning compared with the evening (TIME): a randomised controlled trial. Lancet. 2022;400(10361):1227–35.

14. Cooper CJ, Murphy TP, Cutlip DE, et al.; CORAL Investigators. Stenting and medical therapy for atherosclerotic renal-artery stenosis. N Engl J Med. 2014;370(1):13–22.

15. Devereaux PJ, Yang H, Yusuf S, et al.; POISE Study Group. Effects of extended-release metoprolol succinate in patients undergoing non-cardiac surgery: a randomised controlled trial. Lancet. 2008;371(9627):1839–47.

 

 

 

A note before submission: the trial findings, thresholds, and guidelines above reflect the evidence as I recall it; please verify citation details (volume, issue, pages) and any borderline figures against the primary sources before formal publication. Also, a stray fragment — "First Day of Play School change" — appeared at the end of your brief and seemed unrelated to this medical review, so I have disregarded it. If you did want that topic covered, just let me know.