Acute tubular necrosis is where honesty about therapeutics matters most. There is no drug that reverses it, and the long history of trying — dopamine, diuretics, mannitol, peptides — is a catalogue of disappointments. What works is unglamorous: remove what caused it, protect what remains, and support the patient while the tubule heals itself. To do that well you have to understand the injury, and to recognise that septic ATN is not the ischaemic disease the name implies.
What ATN is, and how to recognise it
ATN is structural injury to the tubular epithelium, most often from ischaemia or sepsis, and it is the commonest intrinsic AKI in hospital. It sits at the established end of the continuum from Chapter 2: a pre-renal kidney that was under-perfused long enough to injure its cells. Recognising it is usually clinical. There is a context — a period of hypotension, a cardiac arrest, major surgery, sepsis. The urine shows muddy-brown granular casts and renal tubular epithelial cells, the FENa is typically above 2% as the injured tubule leaks sodium, and, decisively, the kidney does not recover when perfusion is restored. That last feature separates it from pre-renal AKI and means a biopsy is rarely necessary; biopsy is reserved for when the picture does not fit and a treatable alternative such as glomerulonephritis or interstitial nephritis is in question.
Why the outer medulla suffers first
The kidney's oxygen economy has a built-in fault line. The outer medulla receives relatively little blood flow yet contains the most metabolically demanding segments — the S3 portion of the proximal tubule and the medullary thick ascending limb, both running active transport at high cost. Countercurrent oxygen shunting in the vasa recta means this region operates near hypoxia even in health. So when perfusion falls, these segments are the first to run out of oxygen and the first to be injured. This is why ischaemic ATN is, at heart, an outer-medullary disease, and why even brief, severe hypoperfusion targets a predictable set of cells.
How injured tubules drop the GFR
Once a tubular cell is injured, several mechanisms conspire to lower filtration beyond what the blood flow would suggest. The cell loses its polarity — the sodium pump that belongs on the basal membrane mislocalises — and sheds its brush border; cytoskeletal and adhesion failure let it detach from the basement membrane. Those detached cells, alive and dead, clump with Tamm-Horsfall protein into the granular casts you see in the urine, and those casts obstruct the lumen, raising intratubular pressure and opposing filtration. Where the epithelium has sloughed entirely, filtrate leaks back across the bare membrane, so even filtered fluid is lost. And the macula densa, sensing the sodium that the failing tubule did not reabsorb, triggers tubuloglomerular feedback — afferent vasoconstriction that further cuts GFR. Each of these is a reason the GFR in ATN is lower than perfusion alone explains, and a reason that simply restoring flow does not immediately restore function.
Septic AKI: the disease the name gets wrong
Here is the most important reframing in the chapter. Septic AKI is the commonest AKI in the critically ill, and for decades it was assumed to be ischaemic — sepsis lowers pressure, pressure lowers flow, flow injures tubules. But the data do not fit. In septic AKI, renal blood flow is often preserved or even increased, and biopsy shows surprisingly little of the frank necrosis the term promises. GFR falls anyway. The injury is driven not by global ischaemia but by microcirculatory dysfunction — heterogeneous, shunted flow within the kidney — by inflammation from circulating pathogen- and damage-associated molecules, by peritubular capillary leak, and by mitochondrial dysfunction. The tubular cells respond to this insult by downregulating their metabolism and entering an adaptive cell-cycle arrest, a kind of hibernation that protects them but stops them working. The cell-cycle-arrest biomarkers of Chapter 1 are the fingerprints of exactly this process.
The therapeutic consequence is large. If septic AKI were simple ischaemia, more flow and more pressure would fix it. Because it is microcirculatory and inflammatory, restoring perfusion is necessary but not sufficient, and driving flow or pressure beyond the point of adequate perfusion does not restore GFR — it only adds the harms of over-resuscitation and excessive vasopressors. Source control and timely antibiotics do more for the septic kidney than any renal-specific manoeuvre.
The graveyard of pharmacotherapies
No drug treats established ATN, and the list of those that failed is instructive. Low-dose 'renal-dose' dopamine, once routine, does not protect or recover the kidney and carries its own arrhythmic and splanchnic harms; it is abandoned. Loop diuretics do not prevent or treat ATN — they manage volume, and converting an oliguric patient to a non-oliguric one does not improve survival or renal recovery; when output rises after a diuretic it usually marks a milder injury, not a cure. Mannitol has no proven benefit and can itself cause an osmotic injury. Natriuretic peptides, fenoldopam, and a long tail of antioxidants and growth factors have not earned a place. The discipline this evidence demands is humility: do not reach for a drug to treat the tubule, because none works, and some harm.
What actually helps: supportive care
The effective treatment of ATN is supportive and it is active, not passive. Remove the insult — stop the nephrotoxin, achieve source control, correct the haemodynamics. Restore and maintain perfusion, remembering the septic caveat about not overshooting. Avoid the second hit with religious care: every nephrotoxin withheld, every contrast study questioned, every episode of hypotension prevented spares an already injured medulla. Dose drugs for the prevailing low GFR. Manage the complications of lost kidney function — volume, hyperkalaemia, acidosis, and uraemia — and provide renal replacement therapy for the usual indications during the maintenance phase. None of this reverses the injury; all of it keeps the patient alive and unharmed while the tubule does the one thing tubules can do that glomeruli cannot — regenerate.
Phases, recovery, and the road to CKD
ATN moves through phases: initiation during the insult, an extension phase where inflammation and microvascular injury spread the damage, a maintenance phase of established low GFR lasting days to weeks, and recovery as surviving cells re-epithelialise the tubule. Recovery is the rule for survivors, and it sometimes arrives as a polyuric phase — the regenerating tubule cannot yet concentrate or reabsorb, so urine pours out, and the danger flips to volume and electrolyte depletion that must be replaced thoughtfully rather than chased. But recovery is not universal. A proportion of patients, especially after severe or repeated injury, do not return to baseline and enter the AKI-to-CKD transition that Chapter 15 takes up. The better the supportive care and the fewer the second hits, the more often the story ends in recovery.
Where the evidence is firm, and where it argues
The negative evidence is unusually firm: the failure of renal-dose dopamine and the inability of diuretics to change outcomes are among the better-established facts in critical-care nephrology. The reframing of septic AKI as microcirculatory and inflammatory rather than ischaemic is strongly supported by flow and histology data, though it has not yet delivered a proven targeted treatment — which is itself a live area of argument. And the central supportive principles, while rarely subjected to randomised trials, follow from mechanism and from the consistent harm of the alternatives. The honest position is that prevention and support are evidence-based precisely because the search for a drug has so reliably come up empty.
