Pre-renal Azotaemia

By the end of this chapter you should be able to:

• Distinguish volume depletion (loss of extracellular fluid) from pre-renal azotaemia (the resulting fall in GFR), and explain how one drives the other.
• List the routes of true volume loss — gastrointestinal, renal, skin, third-space, haemorrhagic — and identify the likely source from the history.
• Assess volume status using clinical signs and dynamic measures, and state the reliability of each.
• Separate fluid responsiveness from fluid tolerance, and explain why only about half of hypotensive patients respond to a bolus.
• Choose between a balanced crystalloid and saline, and justify the choice from the chloride mechanism and the trial evidence.
• Prescribe resuscitation in reassessed aliquots, recognise the phases of fluid therapy, and avoid fluid overload.
• Decide when to stop fluid and escalate to a vasopressor, and when a non-responding kidney has crossed into ATN.
• Treat the cause: stop the losses, withdraw the drugs breaking autoregulation, and replace ongoing deficits.

• Volume depletion is a loss of extracellular fluid; pre-renal azotaemia is the fall in GFR it causes when renal perfusion drops — the cause and its renal consequence.
• The parenchyma is intact, so pre-renal AKI reverses the moment perfusion is restored — provided you act before it crosses into ATN.
• Find the leak: gastrointestinal and renal losses are the commonest, but skin, third-space, and haemorrhagic losses all qualify.
• Clinical signs of hypovolaemia are specific when present but insensitive when absent — a normal examination does not exclude depletion.
• Dynamic measures (passive leg raise, pulse-pressure variation, IVC collapsibility) predict fluid responsiveness; static measures like central venous pressure do not.
• Only about half of hypotensive patients are fluid-responsive; giving fluid to a non-responder buys oedema and congestion, not perfusion.
• Ask two questions before every bolus: will it help (responsiveness) and can the patient take it (tolerance)?
• Balanced crystalloids are preferred over 0.9% saline for most resuscitation; the chloride load of saline causes a hyperchloraemic acidosis and renal vasoconstriction.
• Hydroxyethyl starch increases AKI and death and is abandoned; albumin offers no general advantage over crystalloid and harms in traumatic brain injury.
• Resuscitate in 250–500 mL aliquots with reassessment, not fixed large-volume protocols; then optimise, stabilise, and de-escalate.
• Fluid overload is independently linked to higher mortality and worse renal recovery — more fluid is not safer fluid.
• If a patient is not fluid-responsive or is vasodilated, restore perfusion pressure with a vasopressor rather than chasing with volume.
• Treat the cause and stop the offenders: control losses, withdraw NSAIDs and RAAS blockers, and replace ongoing deficits.
• If function fails to recover once perfusion is adequate, reassess for the cross-over to ATN rather than giving still more fluid.

Volume depletion is the commonest reversible cause of acute kidney injury, and the one most often mishandled in both directions — too little fluid in the genuinely dry, too much in the patient who was never going to respond. The skill is not simply giving fluid; it is reading whether perfusion is the problem, whether fluid is the answer, and when to stop. Pre-renal azotaemia rewards early, accurate resuscitation and punishes both neglect and excess.

Volume depletion and pre-renal azotaemia: cause and consequence

Keep the two ideas separate. Volume depletion is a deficit of extracellular fluid — sodium and water lost from the body. Pre-renal azotaemia is what the kidney does about it: as renal perfusion falls, GFR falls, urea and creatinine climb, and the tubule — still structurally intact — conserves sodium and water with everything it has. One is the cause, the other the renal readout. The distinction matters because not every pre-renal state is from true depletion: heart failure, cirrhosis, and sepsis lower the effective arterial blood volume despite a full or overfull body, and they have their own chapters. This chapter is about the patient who has genuinely lost fluid, and the general discipline of restoring perfusion.

Find the leak

Volume leaves by a limited number of routes, and naming the route shortens the work-up. Gastrointestinal losses — vomiting, diarrhoea, nasogastric suction, high-output stoma or fistula — are the everyday culprits. Renal losses come from diuretics, an osmotic diuresis in hyperglycaemia, salt-wasting states, the diuresis that follows relief of obstruction or recovery from ATN, and adrenal insufficiency. The skin and lungs lose fluid through sweating, fever, and burns. Third-spacing sequesters fluid out of the circulation in pancreatitis, bowel obstruction, and the capillary leak of sepsis. And haemorrhage depletes the intravascular compartment directly. The history usually points to the route before any test confirms it.

Reading volume status — and its limits

The bedside signs of hypovolaemia are useful when present and treacherous when absent. A postural fall in blood pressure, a low jugular venous pressure, dry mucous membranes, and delayed capillary refill all support depletion, but their sensitivity is poor — plenty of dry patients look deceptively normal, and skin turgor is unreliable in the elderly. This is why the assessment has moved toward dynamic measures that ask the only question that matters at the bedside: if I give fluid, will the cardiac output rise? A passive leg raise that transiently augments stroke volume, respiratory pulse-pressure variation in a ventilated patient, and a collapsible inferior vena cava on point-of-care ultrasound all predict a response. Static numbers — a single central venous pressure — do not, and should not be used to decide fluid.

Fluid responsiveness is not fluid tolerance

Two separate questions govern every bolus, and conflating them causes harm. The first is responsiveness: on the steep part of the Frank–Starling curve, a preload-dependent ventricle answers a fluid challenge with more output; on the flat part, it does not. Only around half of hypotensive patients sit on the steep part, so half gain nothing from fluid. The second question is tolerance: can the patient accept the volume without congesting the lungs, gut, and kidneys? A patient can be fluid-responsive yet intolerant — their output would rise, but the cost in oedema is unacceptable. Give fluid only when the answer to both is yes, and stop the instant responsiveness disappears.

Choosing the fluid: balanced crystalloid versus saline

For most resuscitation, a balanced crystalloid — a solution whose electrolyte profile approximates plasma — is preferred over 0.9% saline. Saline is misnamed ‘normal’; its supraphysiological chloride load produces a hyperchloraemic metabolic acidosis and, through tubuloglomerular feedback, afferent vasoconstriction that lowers GFR. Large trials in critically ill and emergency-department patients show balanced solutions reduce major adverse kidney events compared with saline. Saline still has its places — a hypochloraemic, hypokalaemic alkalosis from protracted vomiting is the classic indication, and there are nuances in hyponatraemia and brain injury — but the default has shifted to balanced. Colloids have largely lost the argument: hydroxyethyl starch increases AKI and mortality and is abandoned, and albumin confers no general benefit over crystalloid while harming in traumatic brain injury, its specific role in cirrhosis aside.

How much, how fast, and when to stop

Resuscitate in small reassessed aliquots — 250 to 500 mL — rechecking responsiveness and tolerance after each, rather than running fixed large-volume protocols that ignore the individual. It helps to think in phases: an initial resuscitation to rescue perfusion, then optimisation to fine-tune, then stabilisation, and finally evacuation — the deliberate removal of the fluid that resuscitation deposited. The danger at the far end is fluid overload, which is not a benign by-product: a positive cumulative balance is independently associated with higher mortality and worse renal recovery, partly because venous congestion raises renal venous pressure and chokes the perfusion gradient the kidney depends on. More fluid is not safer fluid.

When fluid is the wrong answer

Three situations call for restraint or a different lever. The non-responder gains nothing from more volume and pays in congestion — stop and reassess. The vasodilated patient, as in sepsis, needs perfusion pressure restored with a vasopressor as much as volume; endless litres will not fix a tone problem. And the kidney that fails to recover once perfusion is unequivocally adequate has likely crossed from pre-renal physiology into established ATN, where fluid does nothing for the lesion and everything for the oedema. Recognising these three keeps resuscitation from becoming its own complication.

Treat the cause, withdraw the offenders

Fluid buys time; it does not fix the leak. Stop the losses — antiemetics, control of diarrhoea, correction of hyperglycaemia, transfusion for haemorrhage — and replace ongoing losses measured rather than guessed. At the same time, withdraw what is sabotaging autoregulation: the NSAID blocking the afferent arteriole, the ACE inhibitor or ARB blocking the efferent, the diuretic deepening the deficit. Reversing pre-renal AKI is as much subtraction as addition, a theme inherited directly from the approach chapter.

Table 4.1 — Routes of volume loss

Table 4.2 — Signs of hypovolaemia and their reliability

Table 4.3 — Predicting fluid responsiveness: dynamic beats static

Table 4.4 — Balanced crystalloid versus 0.9% saline

Table 4.5 — Fluid types: role and evidence

Table 4.6 — The phases of fluid therapy

Clinical Reasoning

One triad per mechanism the narrative exposed: the physiology, why it matters, and the bedside move.

Safety & Evidence

Panel A — Red flags

Panel B — Never do

Pitfall 1 — The reflexive bolus

Pitfall 2 — Trusting the CVP

Pitfall 3 — Saline by default

Pitfall 4 — Fluid without an end

Pitfall 5 — Fluid for a tone problem

Graded statements (by evidence type)

Patient Decisions

Natural-frequency estimates for orientation, pooled from trials and cohorts; they vary with population and severity. They show the size of the fluid decisions, expressed as how many of 100 comparable patients are affected.

Apply & Test

Paste-ready notes. Tick the boxes that apply and delete the rest; keep the reasoning visible so the next clinician inherits your assessment, not just your orders.

Template 1 — Volume status assessment and fluid plan

• AKI confirmed (KDIGO stage ___); baseline creatinine ___ ; current ___ .
• Suspected route of loss: ☐ GI ☐ renal ☐ skin/respiratory ☐ third-space ☐ haemorrhage.
• Clinical signs: ☐ postural drop ☐ low JVP ☐ dry mucosae ☐ delayed cap refill (noting limited sensitivity).
• Fluid responsiveness assessed by: ☐ passive leg raise ☐ PPV ☐ IVC — result: ☐ responsive ☐ not responsive.
• Fluid tolerance: ☐ lungs clear ☐ crackles ☐ JVP/congestion present.
• Plan: balanced crystalloid ___ mL aliquot, reassess after each; target ___ ; stop criteria ___ .
• Offenders withdrawn: ☐ NSAID ☐ ACEi/ARB ☐ diuretic; source of loss addressed: ___ .

Template 2 — Post-bolus reassessment

• Aliquot given: ___ mL balanced crystalloid at ___ .
• Responsiveness re-tested: ☐ still responsive ☐ no longer responsive.
• Tolerance: ☐ maintained ☐ new congestion (lungs/JVP/oxygen need).
• Perfusion markers: blood pressure ___ , urine output ___ mL/h, lactate ___ .
• Decision: ☐ continue aliquots ☐ stop and optimise ☐ start/increase vasopressor ☐ begin evacuation.
• If perfusion adequate but function not recovering → reassess for established ATN.