MAP, Shock Index and Pulse Pressure

1. Why Mean Arterial Pressure and Not Systolic

Perfusion of the kidney, the gut and the brain is driven by the mean pressure across the whole cardiac cycle, not by the peak. At a normal rate the heart spends roughly a third of the cycle in systole and two thirds in diastole. That is where the approximation comes from: the diastolic plus a third of the pulse pressure, or equivalently the systolic plus twice the diastolic divided by three. Both forms give the same number.

• The approximation degrades at speed. As the heart rate rises, diastole shortens far more than systole does, so the true mean sits above the calculated one in marked tachycardia.
• A cuff is not an arterial line. Oscillometric devices estimate the mean directly and derive the systolic and diastolic from it, so in a vasoconstricted, hypotensive or arrhythmic patient the mean the machine reports is usually more trustworthy than the systolic beside it.

2. The 65 mmHg Target, and Where It Does Not Apply

The Surviving Sepsis Campaign guideline of 23 March 2026 sets the target, and the direction of travel is towards less pressure, not more, because the higher targets tested cost atrial fibrillation and digital ischaemia without buying survival.

• An initial target of 65 mmHg in septic shock rather than a higher one. A strong recommendation on moderate certainty evidence. Holding exactly 65 is not feasible, so the guideline asks for a range of about 5 mmHg either side.
• For adults aged 65 and above, an initial range of 60 to 65 mmHg. Conditional, on low certainty evidence.

That number is a septic shock target and not a universal floor.

• Long-standing untreated hypertension autoregulates around a higher mean, and such a patient may become oliguric at a mean that would be adequate for someone else. The honest signal there is urine output and mentation.
• Haemorrhage before surgical control runs the other way, towards permissive hypotension.

The mean arterial pressure tells you the pressure; whether it is enough is answered by the lactate, the urine output and the conscious level.

3. The Shock Index Finds the Patient Whose Pressure Has Not Fallen Yet

Tachycardia precedes hypotension by a long way: a fit adult can lose 30 per cent of their blood volume with a systolic pressure still in the normal range. The index crosses 0.9 while the systolic is still reassuring, which is the entire reason for computing it.

• A meta-analysis of adult trauma found an initial index of 1 or above carried a fourfold increase in in-hospital mortality, pooled risk ratio 4.15.
• For massive transfusion, one level 1 centre found the best cut-off at 0.81, sensitivity 85 per cent and specificity 64 per cent, with a negative predictive value of 98 per cent. Above 0.9 is the more widely accepted threshold for activating the protocol.

Treat 0.7 to 0.9 as the band that should make you look again, not the band that reassures. The accepted normal range is 0.5 to 0.7, and some evidence puts the acceptable ceiling at 0.9; pain, fever, anxiety, dehydration and a bronchodilator all put a well ward patient above 0.7. This page reads that band accordingly: in the general adult setting it asks for a second look and a repeat set of vitals, and in the trauma and sepsis settings, where a driver of loss or vasodilatation is already known, it gives the compensated-shock reading.

4. Four Situations Where the Shock Index Lies

• Beta blockade and pacing. The index depends entirely on the tachycardic response to hypovolaemia. Block that response and the index stays normal while the patient bleeds. Not a rare edge case on a ward with a cardiology intake, and the same applies to a paced ventricular rate.
• Age. The heart rate response to a stressor is slower and smaller in an older patient, and chronic hypertension has moved the baseline systolic up. Both push the index down. The age shock index exists because 0.8 means something very different at 75 than at 25.
• Chronic hypertension. A systolic of 120 in someone who normally runs 180 is a 33 per cent fall, and the index computed from it will not look abnormal. Ask what this patient's usual pressure is before deciding what today's means.
• Pregnancy. Physiological tachycardia and a lower systolic shift the whole distribution, so the non-pregnant range of 0.5 to 0.7 is the wrong yardstick. The obstetric range immediately postpartum is 0.7 to 0.9: a value that would be borderline in a non-pregnant adult is normal after delivery, and 1.0 is not.

5. The Pulse Pressure Is a Stroke Volume Proxy

Systolic minus diastolic, conventionally about 40 mmHg in a healthy adult. It reflects the stroke volume ejected against the compliance of the aorta, so a falling pulse pressure in a patient who is otherwise compensating is one of the earliest usable signs that stroke volume is dropping.

• It narrows in hypovolaemia, cardiac tamponade and aortic stenosis.
• It widens with arterial stiffness, aortic regurgitation, thyrotoxicosis, severe anaemia, an arteriovenous fistula and beriberi.
• The proportional pulse pressure is the more useful of the two, because it removes the dependence on the absolute pressure: the pulse pressure divided by the systolic, as a percentage.

Stevenson and Perloff established the bedside haemodynamic profiles in heart failure. In 50 patients with known chronic heart failure a proportional pulse pressure below 25 per cent identified a cardiac index below 2.2 L/min/m², sensitivity 91 per cent and specificity 83 per cent. A later series using a cut-off of 27.4 per cent reported 95.5 and 90.1 per cent. In a breathless patient with a low output state, that single division is the closest thing to a cardiac index available without a catheter.

6. Reading the Three Together

7. Indian Practice Notes

• Postpartum haemorrhage, still a leading direct cause of maternal death. The shock index is promoted here because it needs a watch and a cuff rather than a laboratory. A value of 0.9 or above in the immediate postpartum period identified every woman requiring intensive care in the series that proposed it, at the cost of a low specificity. For a screening threshold in a district hospital that is the right trade.
• Dengue. The narrowing pulse pressure of shock in severe dengue is the taught sign: 20 mmHg or less with a rising haematocrit is the moment for fluid, well before the systolic falls. That grade, once taught as dengue shock syndrome, was retired by the National Guidelines for Clinical Management of Dengue Fever 2023, which classify into Groups A, B and C instead. Neither the sign nor its timing is affected by the change of name, and the shock index and pulse pressure move hours earlier.
• Snakebite and organophosphorus poisoning, both of which distort the picture. Envenoming can cause capillary leak with a widening then collapsing pulse pressure; organophosphates produce bradycardia, so the shock index in a poisoned patient may be low while the patient is profoundly unwell.
• The equipment. An oscillometric monitor in a hypotensive, vasoconstricted or atrially fibrillating patient can read the systolic several tens of millimetres away from an intra-arterial trace, and the direction of that error is not predictable. Where the cuff and the patient disagree, believe the patient, check the cuff size, and repeat it manually.

8. What This Tool Deliberately Does Not Do

• It does not diagnose shock, which is a clinical state defined by inadequate tissue perfusion and not by any ratio.
• It does not compute cerebral perfusion pressure, which needs an intracranial pressure it cannot know.
• It does not give a fluid volume. That belongs with the fluid module and with an assessment of fluid responsiveness, which a single blood pressure cannot provide.
• It does not cover children, in whom the shock index thresholds are age-banded and different.
• It computes nothing from a trend, because it is given one reading. Two readings twenty minutes apart, read side by side, will tell you more than anything on this page.