tuckermedical.comMedical education2026

Renal function · Lab interpretation

Two ways to measure a kidney

Every eGFR we act on is an estimate built from a blood marker the kidney happens to clear. Creatinine has done that job for 70 years. Cystatin C answers a different set of questions — and increasingly, it answers them better.

The immediate reason this deck exists: a steady stream of otherwise healthy people arriving with an "abnormal" creatinine and a flagged eGFR, because they started taking creatine.

Format
~20 min + Q&A

Audience
Physicians, primary care and concierge practice, nursing and clinical support

Includes
Live CKD-EPI 2021 calculator (slide 13)

AuthorSteven Tucker, M.D. (USA), FACP (USA), FAMS (Singapore)
Senior Consultant, Medical Oncology
Founder and Medical Director, Tucker Medical, Singapore

The impetusA supplement, not a disease

The most common cause of an "abnormal" creatinine we now see isn't kidney disease

Creatine monohydrate has moved from the weight room into general use — athletes, the longevity and healthspan community, and increasingly older adults taking it for sarcopenia and cognition. It is one of the best-evidenced supplements available, and it reliably raises serum creatinine without touching the kidney. Most people taking it have no idea, and neither does the lab.

Why the number moves

Creatinine is the non-enzymatic breakdown product of creatine and creatine phosphate — roughly 1–2% of the total pool converts per day, irreversibly. Supplementing enlarges the pool, so daily creatinine production rises proportionally. Nothing has happened to filtration; the numerator simply got bigger.

Input	Typical dose	Effect on serum creatinine
Creatine — loading phase	20 g/day × 5–7 d	↑ largest effect; can reach 20–30% at the extreme
Creatine — maintenance	3–5 g/day	↑ modest, sustained
L-arginine	varies	↑ — upstream substrate for creatine synthesis
High-protein / high-meat diet	≥1.2 g/kg/day	↑ additive
Cooked-meat meal	single	↑ transient, hours

A documented case report shows the same false rise from L-arginine alone, with cystatin C unchanged throughout and creatinine normalising a week after stopping — arginine is the rate-limiting substrate for endogenous creatine synthesis. The arginine evidence rests on that single report and should be treated as plausible rather than established.

Keep the magnitude in proportion. The 20–30% figures come from loading-phase case reports and are an upper bound, not the typical picture. Pooled randomised data are more modest: two 2025–2026 meta-analyses found a small increase in serum creatinine on the order of 0.07–0.12 mg/dL (roughly 6–11 µmol/L), largest in trials of a week or less, with no significant change in urea or eGFR and no signal of renal injury. The clinical problem is therefore rarely the size of the rise — it is that a small rise crosses a reporting threshold and triggers a cascade.

The cleanest single dataset

Gualano et al., Am J Kidney Dis 2010 — a 20-year-old man with a single kidney, on 2.8 g/kg/day protein, given creatine 20 g/day × 5 d then 5 g/day × 30 d, with GFR measured by ⁵¹Cr-EDTA clearance:

Measure	Pre	Post
Serum creatinine (mg/dL)	1.03	1.27
Estimated CrCl^†	88	71
Measured GFR — ⁵¹Cr-EDTA	81.6	82.0
Proteinuria (mg/d)	130	120
Albuminuria (mg/d)	4.6	2.9

The estimate fell by 17 points. True filtration did not move at all. Albuminuria — the one marker that reflects actual glomerular integrity — improved.

^†Reported in the original paper as mL/min/1.73 m². Creatinine clearance is conventionally expressed in mL/min without indexing; the distinction matters and is the subject of slide 12.

Two failure modes follow, and both are common. A healthy person gets labelled with CKD, worked up, and told to stop a supplement that was helping them. Or the number is waved away as "just the creatine" — and a genuine early decline is missed.

The resolution is one test. Cystatin C production is unaffected by creatine, protein intake, or muscle mass. If the eGFR_cys is normal, the creatinine rise was pool expansion. If it isn't, you have found real disease in someone you would otherwise have reassured. Either way you now know — which is the entire argument of this deck.

Clinical stakeseGFR drives real decisions

The eGFR isn't a number on a report. It's a gate.

Before we argue about which marker is better, be clear about what the number actually controls in this practice.

Decisions gated by eGFR

Carboplatin dose — Calvert formula uses GFR directly. Wrong GFR = wrong dose, every cycle.
Cisplatin eligibility — usually needs CrCl ≥ 60 mL/min.
Iodinated CT contrast — risk conversation below eGFR 30.
Gadolinium agent selection for MRI.
Drug dose adjustment — DOACs, metformin, antivirals, gabapentin, LMWH, many oral targeted agents.
CKD staging & nephrology referral.
Longevity / risk panels — eGFR is one of the strongest predictors of cardiovascular and all-cause mortality we measure.

What "eGFR" is really telling you

GFR = the volume of plasma the glomeruli clear per minute. We can't measure it routinely, so we infer it from a substance in the blood whose level rises when filtration falls.

That inference only holds if the body produces the marker at a steady, predictable rate. Everything that follows in this session is about the ways that assumption breaks.

The core idea: a rising marker level can mean the kidney is filtering less — or that the body is simply making more of it. The marker cannot tell you which.

How imprecise is it? For a creatinine-based eGFR reported as 60, the 95% confidence interval for that individual's true GFR runs from roughly 35 to 85 mL/min/1.73 m² (Shafi et al., Ann Intern Med 2022). That single number spans CKD stage G3b to normal. The lowercase "e" is doing an enormous amount of work.

Reported eGFR is indexed to 1.73 m² body surface area. That indexing is right for staging CKD and wrong for dosing drugs. Slide 12.

First principlesFive criteria

What a perfect filtration marker would look like

The ideal marker is produced at a constant rate, gets filtered at the glomerulus, and is then left completely alone by the rest of the nephron. Nothing in human biology does all five. Creatinine and cystatin C each fail different ones.

Watch the two markers separate. Both are filtered. The green particles are pulled out of the proximal tubule and destroyed, so none appears in urine. The red particles pass all the way through — and are joined mid-tubule by extra creatinine secreted straight from the capillary, clearance that never involved filtration at all.

01

Produced at a constant rate

Output shouldn't swing with diet, muscle, steroids, thyroid or inflammation.

02

Freely filtered at the glomerulus

Small, unbound, passes the filtration barrier without resistance. Both pass

03

Not reabsorbed downstream

Anything pulled back into blood makes filtration look worse than it is. Both pass — with an asterisk

04

Not secreted into the tubule

Extra secretion clears marker without filtration, so GFR looks better than it is. Creatinine fails

05

Not affected by non-kidney disease

Level should move for renal reasons only. Cystatin C fails

Cystatin C is reabsorbed and fully catabolised in the proximal tubule — none returns to blood, so serum levels stay valid. The trade-off: it never reaches urine, so you cannot do a cystatin C clearance study.

Marker profileThe classic · in use since the 1950s

Creatinine

A waste product of creatine phosphate turnover in skeletal muscle. Production is proportional to muscle mass — which is exactly the problem.

Where it comes from

Non-enzymatic breakdown of creatine/creatine phosphate in muscle, ~1–2% of the muscle pool per day
Dietary contribution from cooked meat (heat converts creatine → creatinine)
Creatine supplements add to the pool

How the kidney handles it

Freely filtered ✓
Not reabsorbed ✓
Actively secreted by proximal tubule OCT2 / MATE transporters ✗
Secretion is ~10–15% of clearance normally, rising toward 30–40% in advanced CKD — so CrCl systematically overestimates true GFR as function declines

Why we still use it

Cheap, universal, on every basic panel
Decades of outcome data and drug labels built on it
Enzymatic assays are IDMS-standardised across labs
Fast turnaround, no special handling

Older Jaffe (picrate) assays are prone to interference from bilirubin, glucose, ketones and some cephalosporins. Enzymatic assays are preferred and are what most Singapore labs now run.

Rule of thumb: high muscle mass → high creatinine → eGFR underestimated. Low muscle mass (sarcopenia, cachexia, amputation, paralysis, cirrhosis, frailty, prolonged bed rest) → low creatinine → eGFR overestimated, sometimes dramatically. The second error is the dangerous one, because it lets us dose drugs a kidney can't handle.

Marker profileClinically standardised since 2010

Cystatin C

A 13 kDa cysteine protease inhibitor produced at a steady rate by all nucleated cells. Its production is independent of muscle — which is exactly its advantage.

Where it comes from

Housekeeping gene CST3, expressed by every nucleated cell
Production largely constant across body size and muscle mass
Not affected by meals, meat intake or supplements

How the kidney handles it

Freely filtered ✓
Not secreted ✓
Reabsorbed by proximal tubule and completely metabolised there — nothing returns to blood, so serum level remains a clean filtration signal
Consequence: no urinary cystatin C clearance test exists

Practical notes

Immunoassay (PENIA/PETIA), calibrated to the international ERM-DA471/IFCC reference material
Typical adult range ≈ 0.6–1.0 mg/L (confirm against your lab's interval)
No fasting needed; ordinary serum tube; stable at room temperature
More expensive than creatinine and not part of a routine panel — it has to be requested deliberately

The single most useful property: cystatin C is blind to muscle. In a cachectic patient with a "beautiful" creatinine of 50 µmol/L, cystatin C is the marker that tells you the truth.

Head to headScorecard against the five criteria

Side by side

Property	Creatinine	Cystatin C	Verdict
Freely filtered	✓ Yes	✓ Yes	Tie
Not reabsorbed into blood	✓ Yes	✓ Reabsorbed but fully catabolised — none returns	Tie
Not secreted	✗ Secreted; increases in early CKD and blocked by several drugs	✓ Not secreted	Cystatin C
Independent of muscle mass	✗ Directly proportional to muscle	✓ Independent	Cystatin C
Independent of non-renal disease	✓ Relatively	✗ Rises with steroids, thyroid disease, inflammation, adiposity, malignancy	Creatinine
Diet-independent	✗ Cooked meat, creatine supplements	✓ Yes	Cystatin C
Measurable in urine (clearance study)	✓ 24-h CrCl possible	✗ Catabolised — never reaches urine	Creatinine
Cost & availability	✓ Cents, on every panel	✗ Higher cost, must be requested	Creatinine
Assay standardisation	✓ IDMS-traceable	✓ ERM-DA471 since 2010	Tie
Predicts mortality & CV events	Good	Better — consistently stronger association, even within the "normal" creatinine range	Cystatin C

Neither wins outright. They fail on opposite axes, which is why the combined equation using both markers is more accurate than either one alone — the errors partly cancel.

One correction if you've seen the popular infographic on this topic. Several widely-shared versions tick "not reabsorbed" for cystatin C without qualification. That is an oversimplification: cystatin C is reabsorbed — essentially 100% of it — by the proximal tubule, and then completely catabolised there. The reason it still works as a filtration marker is that none of it returns to the circulation intact, so serum levels remain a clean signal. The practical consequence of getting this right: there is no such thing as a urinary cystatin C clearance test.

Error sourcesCreatinine

What fools creatinine

Read the last column. That is the direction the reported eGFR moves — and whether the error is the safe kind or the dangerous kind.

Condition	Effect on serum creatinine	Reported eGFR	Clinical risk
Sarcopenia, cachexia, advanced cancer	↓ Low	Over-estimated	Overdosing renally cleared drugs
Amputation, paraplegia, prolonged immobility	↓ Low	Over-estimated	Same
Cirrhosis (reduced hepatic creatine synthesis)	↓ Low	Over-estimated	Masks hepatorenal decline
Frail elderly, low protein intake, vegan diet	↓ Low	Over-estimated	Same
Bodybuilder, high muscle mass, creatine supplement	↑ High	Under-estimated	False CKD label; unnecessary dose reduction
Large cooked-meat meal before the draw	↑ High (transient, hours)	Under-estimated	Spurious "worsening renal function"
Drugs blocking tubular secretion (next slide)	↑ High	Under-estimated	Pseudo-AKI — treatment stopped unnecessarily
Advancing CKD (secretion fraction rises)	Relatively lower than expected	Over-estimated	Under-recognised severity
Ketoacidosis, marked hyperbilirubinaemia (Jaffe assay)	↑ Artefact	Under-estimated	Assay interference, not physiology

Watch for this pattern: creatinine drifting down over cycles of chemotherapy while the patient visibly loses weight and muscle. That falling creatinine is not improving kidney function.

Error sourcesCystatin C

What fools cystatin C

Cystatin C's weakness is the mirror image: production is not driven by muscle, but it is nudged by inflammation, steroids and metabolic state. Several of these are routine in an oncology clinic.

Condition	Effect on serum cystatin C	Reported eGFR	Note
Corticosteroids (dexamethasone, prednisolone)	↑ High, dose-dependent	Under-estimated	Highly relevant — dexamethasone is standard chemo premedication
Hyperthyroidism	↑ High	Under-estimated	Long-taught, from older observational work. Magnitude now debated — thyroid hormone analogues were not significant after full adjustment in the 2026 iohexol-validated data
Hypothyroidism	↓ Low	Over-estimated	Same caveat. Still worth checking TFTs when a result is unexplained
Opioids	↑ High	Under-estimated	Newly confirmed independent of mGFR; effect small and probably not clinically meaningful
Loop diuretics	↑ High	Under-estimated	Same — small, independent of mGFR
Systemic inflammation, high CRP, sepsis	↑ High	Under-estimated	Partly a real marker of illness, not pure artefact
Active malignancy (some tumour types)	↑ High	Under-estimated	Relevant to our whole caseload
Obesity / high adiposity	↑ High	Under-estimated	Adipose tissue contributes to production
Current smoking	↑ High	Under-estimated	Modest effect
Diabetes mellitus	↑ High	Under-estimated	≈8.5% higher after adjustment for GFR; independent of glycaemic control, mechanism unclear
HIV infection	↑ High	Under-estimated	Reported in the older literature; effect size modest

Timing matters. If a patient received IV dexamethasone yesterday, a cystatin C drawn today may read falsely high. Where the result will change management, note steroid exposure on the request and, if practical, draw before the steroid course or ≥1–2 weeks after it ends.

What does not affect it: muscle mass, meals, meat intake, creatine supplements, amputation, immobility, tubular secretion drugs. That is the entire reason we order it.

Best current evidence · Russel et al., JASN 2026 Much of the confounder list above comes from older observational work. The strongest test of it to date used iohexol-measured GFR in 5,595 adults and asked which of eight drug classes move cystatin C once true filtration and illness are accounted for. Only three survived: systemic glucocorticoids, opioids and loop diuretics. For everything except glucocorticoids the increases were small and likely not clinically meaningful — and thyroid hormone analogues and immunosuppressants were not significant after full adjustment. The most useful finding for practice: in patients taking any of these drugs, eGFR_cr-cys was consistently more accurate than either marker alone. The answer to a confounded cystatin C is not to abandon it — it is to use both markers together.

PharmacologyCreatinine rise without GFR change

Pseudo-AKI: drugs that raise creatinine without touching the kidney

These agents block the proximal tubule transporters (OCT2, MATE1, MATE2-K) that secrete creatinine. Serum creatinine rises within days, plateaus, and reverses on stopping. True GFR is unchanged — cystatin C stays flat, which is how you prove it.

Common offenders

Trimethoprim (± sulfamethoxazole)	Classic; rise of 10–30%
Cimetidine	Historically used deliberately to block secretion
Dolutegravir, bictegravir	HIV integrase inhibitors
Cobicistat, ritonavir	Pharmacokinetic boosters
Abemaciclib, ribociclib	CDK4/6 inhibitors — we see this in breast cancer patients
Tepotinib, capmatinib	MET inhibitors
Vandetanib	Multikinase inhibitor; OCT2/MATE inhibition
Salicylates (high dose)	Reduce tubular secretion
Pyrimethamine	Same mechanism
Fenofibrate	Different mechanism (↑ production) but same trap

Piperacillin–tazobactam is often listed here too. The picture is mixed: it interferes with creatinine handling and is genuinely associated with AKI when combined with vancomycin. Do not dismiss a rise on pip-tazo as artefact.

The pattern that identifies it

Creatinine rises within 1–14 days of starting the drug
Rise is modest (typically <30%) and then plateaus
Urine output, electrolytes and urinalysis are unremarkable
No new hypotension, obstruction or nephrotoxin
Cystatin C is unchanged ← the confirmatory test
Reverses within days of stopping

Why this matters to us: a patient on abemaciclib whose creatinine rises 20% may have their effective cancer therapy held or dose-reduced for no reason. A single cystatin C settles it. Flag this pattern to the doctor rather than assuming AKI.

EquationsCKD-EPI 2021 · race-free

Which equation the lab is using

Since 2021 the standard is the race-free CKD-EPI set (Inker et al., NEJM 2021). The old 2009 equation applied a multiplier for Black race; that coefficient has been removed because race is a social, not biological, variable and the adjustment systematically delayed transplant referral and CKD diagnosis. Three flavours exist.

eGFR_cr

Creatinine + age + sex.

Default on every panel. Accurate enough for population screening, unreliable at the extremes of muscle mass.

eGFR_cys

Cystatin C + age + sex. No race coefficient — but note it does retain a sex coefficient (×0.932 for women). What it drops is not sex, but the reliance on muscle mass as the thing sex and age are standing in for.

Use when creatinine is untrustworthy — or as the confirmatory test. A cystatin C equation free of both race and sex has since been published (Pottel et al., NEJM 2023) and may become the standard.

eGFR_cr-cys

Both markers together.

The most accurate of the three against measured GFR, in essentially every validation cohort. This is the one to prefer when you have both values.

Legacy equations you'll still see

Cockcroft–Gault	1976. Estimates creatinine clearance, not GFR, in mL/min (not indexed). Still specified in many drug package inserts, so pharmacy may use it deliberately.
MDRD	Superseded. Inaccurate above eGFR 60; many labs used to report ">60" rather than a number because of this.
CKD-EPI 2009	Race-based. Being phased out.

Practical pointBefore you conclude a patient has deteriorated, check which equation and which units the lab used. Switching between MDRD, CKD-EPI 2009 and CKD-EPI 2021 can move the reported number by 5–10 mL/min/1.73 m² in a patient whose kidneys have not changed at all. This is one of the most common sources of a false "declining renal function" conversation — particularly when results arrive from a second lab, another country, or a health-screening package.

TrapsUnits · indexing · staging

Two unit traps that change doses

Trap 1 — µmol/L vs mg/dL

Most of the world reports creatinine in µmol/L. The United States, most online calculators and many drug labels use mg/dL.

mg/dL = µmol/L ÷ 88.4

µmol/L	mg/dL
50	0.57
70	0.79
90	1.02
120	1.36
200	2.26

Cystatin C is reported in mg/L everywhere — no conversion needed.

CKD stage by eGFR

G1	≥ 90	Normal or high
G2	60–89	Mildly decreased
G3a	45–59	Mild–moderate
G3b	30–44	Moderate–severe
G4	15–29	Severe
G5	< 15	Kidney failure

G1 and G2 are only CKD if another marker of kidney damage is present. Staging is incomplete without albuminuria.
SI: A1 <3 · A2 3–30 · A3 >30 mg/mmol
US: A1 <30 · A2 30–300 · A3 >300 mg/g
Convert: mg/mmol × 8.84 ≈ mg/g. Slide 20.

Trap 2 — indexed vs absolute GFR

This is the one that actually harms people, and almost nobody checks it.

The eGFR on the report is mL/min/1.73 m² — normalised to the body surface area of an average adult. That normalisation exists so populations can be compared and CKD staged consistently. It is correct for those purposes.

Drug dosing does not want a normalised number. It wants this patient's actual clearance, in mL/min:

De-index before dosing Absolute GFR (mL/min) = reported eGFR × (BSA ÷ 1.73)

Patient	BSA	Reported eGFR	Actual clearance	Error if you skip this
45 kg · 150 cm	1.37	60	48	+25% overdose
52 kg · 158 cm	1.51	60	52	+14% overdose
70 kg · 170 cm	1.81	60	63	−4%, negligible
105 kg · 183 cm	2.27	60	79	−24% underdose

The direction of the error depends entirely on body size. Small patients get too much drug; large patients get too little. Both matter, but the first is the one that lands someone in hospital.

Where this bites hardest: the Calvert formula for carboplatin, which takes GFR as a direct input. Also vancomycin, DOACs, aminoglycosides, and any agent with a narrow therapeutic index in a patient at the extremes of body size. See slide 13 to watch the numbers move, and slide 17 for the carboplatin case.

InteractiveCKD-EPI 2021 · all three equations

See the discordance yourself

Default case: a 62-year-old woman, twelve months into treatment, visibly sarcopenic. Her creatinine looks excellent. Change the numbers and watch the three estimates separate.

Age (years)

Sex

Creatinine

Creatinine units

Cystatin C (mg/L)

Height (cm) — optional

Weight (kg) — optional

eGFR_crCreatinine only

—

eGFR_cysCystatin C only

—

eGFR_cr-cysBoth · most accurate

—

Absolute GFR (de-indexed)

—

DuBois BSA. This is the number drug dosing needs — see slide 12.

Calvert carboplatin, AUC 5

—

Dose = AUC × (GFR + 25), GFR left uncapped here so the effect is visible. Illustration only — not a prescribing tool.

InterpretationWhen the two markers disagree

Discordance is information, not noise

Define eGFR_diff = eGFR_cys − eGFR_cr. A large negative value — cystatin C much worse than creatinine — is one of the more informative findings on a routine panel.

How to read the gap

Pattern	Usual explanation	Action
Within ~10 mL/min	Concordant	Trust the combined equation
eGFR_cys much lower	Low muscle mass, inflammation, steroids, or true occult CKD	Use combined or measured GFR before dosing
eGFR_cys much higher	High muscle mass, creatine supplement, secretion-blocking drug, recent meat meal	Suspect pseudo-AKI; check drug list

Shrunken pore syndrome: a persistent eGFR_cys/eGFR_cr ratio below ~0.6–0.7 with no other explanation. Thought to reflect narrowed glomerular pores that hold back mid-sized molecules while small ones pass freely. Independently associated with higher cardiovascular and all-cause mortality.

Why the gap predicts outcomes

A large negative eGFR_diff tracks with sarcopenia, frailty and inflammatory burden — so it captures both kidney and whole-body reserve
In cohort studies it is associated with higher mortality, more heart failure and more adverse drug events, independent of eGFR itself
The creatinine-to-cystatin C ratio (CCR) has been validated as a bedside "sarcopenia index" — see the Singapore cohort data below

Framing for the longevity side of the practice: two people can both have an eGFR of 90 on creatinine. If one has a cystatin C-based eGFR of 60, they are not the same patient, and the difference is visible years before creatinine moves.

The same discordance, pointed at muscle instead of kidney. Because creatinine production scales with muscle and cystatin C does not, their ratio is a filtration-independent index of skeletal muscle mass — the creatinine-to-cystatin C ratio, or CCR. That has become a practical sarcopenia marker, and the strongest longitudinal data come from a Singapore cohort. Next slide.

Special populationMenopause · body composition · MHT

Menopause, MHT and the two markers

The menopausal transition changes body composition in exactly the way that corrupts a creatinine-based eGFR — and menopausal hormone therapy changes it again. This is the largest population in which the errors in this deck apply, and the one where they are least often considered.

What the transition does

Falling oestrogen drives increased visceral fat and reduced muscle mass — changes BMI does not capture. Prof Yong's group notes a large proportion of Singaporean women are "skinny fat": normal BMI, high visceral fat, low muscle.

They call it the Janus-like effect, after the god of transitions: the same passage that ends reproductive life opens a metabolic one.

The consequence for this deck: a woman losing muscle across the transition has a falling creatinine and therefore a rising eGFR_cr. It is the identical false reassurance as the cachectic oncology patient on slide 18 — in a population perhaps a thousand times larger, and almost never flagged.

The Singapore evidence — IWHP

1,200 Chinese, Malay and Indian women aged 45–69, recruited 2014–16; ~70% post-menopausal at baseline, ~90% by follow-up. In the CCR analysis: 891 women, mean age 56.2, followed 6.6 years with MRI-measured muscle volumes.

Lowest CCR tertile (<8.16)	Effect
Thigh fat-free muscle volume	−0.35 L
Narrow gait speed	−0.049 m/s
Usual gait speed	−0.029 m/s
Grip, chair stands, one-leg stand	no association

Muscle volume and gait speed are both sarcopenia criteria — so a blood test taken years earlier flags the trajectory before any performance test does.

Companion IWHP diabetes data: high visceral fat alone 1.78× risk of prediabetes/T2D; weak strength plus high visceral fat 2.63×; within the weak-strength group, high vs low visceral fat 2.84×.

What MHT itself does to the markers

Sex steroids move both analytes. The cleanest experimental evidence comes from gender-affirming hormone therapy, where exposure is defined and prospective:

Exposure	Creatinine	Reads as
Oestradiol (+ antiandrogen)	↓ ~0.065 mg/dL at 12 mo	eGFR_cr higher
Testosterone	↑ ~0.131 mg/dL at 12 mo	eGFR_cr lower

The testosterone signal is muscle accrual, not renal injury — meta-analysis attributes the eGFR dip to increased creatinine production, with values stabilising by 18–24 months. Cystatin C rose ~0.052 mg/L on testosterone, worth about 6 mL/min/1.73 m² on eGFR_cys.

Do not transfer the numbers. GAHT doses are not MHT doses. Treat these as direction and mechanism, not magnitude — but note that testosterone is increasingly used in women's MHT protocols, and it will move creatinine.

Does MHT help or harm the kidney? Honestly, unsettled

Supports benefit	Longer lifetime oestradiol exposure — earlier menarche, later menopause — is associated with lower CKD risk. A 2018 meta-analysis found postmenopausal hormone therapy associated with reduced albuminuria, and estradiol therapy has been linked to lower risk of kidney failure.
Supports caution	Older observational work in elderly women reported that oral oestrogen was associated with faster eGFR loss. That analysis used the MDRD equation on creatinine alone — the weakest available instrument, and precisely the one confounded by the muscle changes under discussion.
Net	Do not use MHT status either to reassure or to alarm about renal function. Measure it properly instead.

What to do in an MHT clinic

Baseline creatinine and cystatin C before starting, in any woman where you intend to follow renal function or body composition over time. Without the baseline pair, later changes are uninterpretable.
If you add testosterone, expect creatinine to rise. Check cystatin C before working it up as AKI or reducing a dose — this is the pseudo-AKI pattern of slide 10 arriving by a different route.
Track CCR as a response marker to what you are actually prescribing alongside MHT: resistance training and adequate protein. It moves with the intervention in a way eGFR does not.
Always pair with urine ACR. Albuminuria is where the MHT evidence is strongest and where the outcome data sit — see slide 20.
Do not stage CKD on eGFR_cr alone in a woman whose weight or body composition is shifting.

Ordering guidanceAligned to KDIGO 2024 CKD

When to add cystatin C

KDIGO's 2024 CKD guideline recommends confirmatory testing when creatinine-based eGFR is likely to be inaccurate and the result will change what you do. Where both markers are available, the combined equation is preferred. Below is our working trigger list.

Order it when…

The patient takes creatine, L-arginine or a very high-protein diet — and has an otherwise unexplained creatinine or eGFR flag
Body composition is atypical — cachexia, sarcopenia, BMI <18.5 or >35, bodybuilder, amputee, paraplegia
Before carboplatin or cisplatin in any patient who has lost significant weight
eGFR sits at a decision threshold — 30, 45 or 60 — and the number decides therapy
Unexplained creatinine rise on a secretion-blocking drug (abemaciclib, ribociclib, trimethoprim, dolutegravir)
Cirrhosis or advanced liver disease
Before a nephrology referral or a transplant workup
Executive / longevity panels where the goal is early detection rather than staging
Midlife women through the menopausal transition, and before starting MHT — where the ratio doubles as an early sarcopenia signal and testosterone will move creatinine (slide 15)
Discordance between eGFR and the clinical picture

Think twice when…

The patient is on high-dose corticosteroids right now
There is active untreated thyroid disease
The patient is acutely septic or has florid inflammation — the result is valid but harder to attribute
Nothing about the answer would change management

The creatine patient — two ways to settle it

Wash out and repeat	Stop creatine and hold high-protein loads, then repeat creatinine no sooner than 4 weeks later. The intramuscular pool takes weeks to normalise, so a 2–3 day break is not enough. Cheap, but slow, and asks the patient to stop something beneficial.
Add cystatin C	No washout, no behaviour change, answers the question on the same draw. Preferred when the patient wants to continue, when a dose decision is pending, or when they are anxious.

Either way, pair it with a urine ACR. In the Gualano case the creatinine rose while albuminuria fell — the ACR is what distinguishes a bigger creatinine pool from a leaking glomerulus. Also check that urea and electrolytes are unremarkable; an isolated creatinine rise with a normal urea is the supplement pattern.

Before you rely on it

Cystatin C goes in a standard serum tube alongside the routine renal panel — no separate visit, no fasting. Add a clinical note on the request: creatine or supplement use, weight loss, steroid exposure, or suspected pseudo-AKI.

Confirm with your own reference lab what cystatin C costs, its turnaround time, whether it is covered, and — importantly — whether the lab returns eGFR_cys and eGFR_cr-cys or only the raw value. Many return only the raw value, which is not directly interpretable.

Oncology applicationCarboplatin & cisplatin

Where a wrong GFR becomes a wrong dose

Carboplatin — the Calvert formula

Dose (mg) = target AUC × (GFR + 25)

Unlike almost every other cytotoxic, carboplatin is dosed off renal function directly rather than body surface area. Three things go wrong routinely:

Using indexed eGFR. Calvert needs absolute GFR in mL/min. Slide 12.
Using creatinine in a sarcopenic patient. Overestimated GFR → overdose → deeper myelosuppression, more febrile neutropenia, more dose delays.
Not capping GFR. With IDMS-standardised creatinine assays, estimated GFR can run falsely high. The familiar advice to cap GFR at 125 mL/min is specifically an FDA-era recommendation for Cockcroft–Gault-derived clearance, not a universal rule — see the note below.

In the default calculator case on slide 13, dosing from creatinine gives roughly 540 mg; dosing from the combined equation gives roughly 415 mg. Same patient, same day — a ~25% difference in exposure across every cycle.

Guidance is moving. The international ADDIKD consensus (2025) recommends directly measured GFR as the preferred input for anticancer dosing in kidney dysfunction, with BSA-adjusted eGFR as the alternative, and does not frame a fixed 125 mL/min cap the way earlier US guidance did. If your unit still applies the cap reflexively, it is worth checking that against ADDIKD before treating it as current standard.

Cisplatin

Typically requires CrCl ≥ 60 mL/min to proceed at full dose
Directly nephrotoxic — proximal tubular injury, magnesium and potassium wasting
Requires pre- and post-hydration; monitor Mg²⁺ and K⁺ every cycle
A creatinine that looks stable in a wasting patient can conceal real cumulative injury

Other renally-sensitive agents

High-dose methotrexate — delayed clearance is life-threatening; needs accurate baseline GFR, urine alkalinisation, level monitoring
Pemetrexed — contraindicated below CrCl 45
Bisphosphonates (zoledronic acid) — dose-adjusted by CrCl; label advises against use below CrCl 35
Capecitabine — dose-reduce at CrCl 30–50

InterpretationWhen both markers mislead at once

The steroid trap — and why some patients break both markers together

Most of this deck treats the two markers as alternatives: when one is unreliable, reach for the other. There is a large group of patients in whom both are unreliable simultaneously, and the errors point in opposite directions. Recognising the pattern matters more than memorising any threshold.

Falling muscle mass

Weight loss, deconditioning, frailty, chronic illness, prolonged steroid exposure itself (steroid myopathy) — all reduce creatinine production.

Effect: eGFR_cr falsely high. Risk of overdosing renally cleared drugs.

Corticosteroids & inflammation

Glucocorticoids increase cystatin C production in a dose-dependent way. Systemic inflammation and high CRP do the same.

Effect: eGFR_cys falsely low. Risk of withholding treatment or over-restricting.

What to do

Use the combined equation as the default. Note steroid exposure on the request. Where a high-stakes dose decision rests on the number and the two markers disagree by more than about 20 mL/min/1.73 m², consider measured GFR.

Who this describes — it is not a niche group

Setting	Steroid exposure	Muscle loss
Cancer on systemic therapy	Dexamethasone premedication, most cycles	Cachexia, treatment-related sarcopenia
Polymyalgia rheumatica, vasculitis, RA	Long-term prednisolone	Steroid myopathy, age, reduced activity
COPD, severe asthma	Repeated oral bursts	Deconditioning, low BMI
Solid-organ transplant	Maintenance steroids	Post-transplant loss, plus real CKD to detect
Inflammatory bowel disease	Induction courses	Malabsorption, low lean mass
Frail elderly on any steroid	Variable	Age-related sarcopenia

For general practice and longevity medicine, the last three rows are the ones you will meet most often. A 76-year-old on prednisolone for PMR who has lost 6 kg is the canonical case.

A worked example

Timepoint	Weight	Creatinine	eGFR_cr	Cystatin C	eGFR_cys
Baseline	62 kg	78 µmol/L	78	0.95 mg/L	75
6 months	54 kg	68 µmol/L	90	1.15 mg/L	60
12 months	48 kg	62 µmol/L	98	1.35 mg/L	48

Creatinine "improved" by 16 µmol/L and eGFR_cr rose 20 points. Over the same year the patient lost 14 kg and true filtration fell by roughly a third.

The pattern in one line: in a patient who is losing weight, an improving creatinine is a red flag, not a reassurance.

ImagingCT contrast · gadolinium

Renal screening before imaging

Few areas have shifted as much in the last decade, and few have as much outdated practice still in circulation. Know the thresholds and the reasoning behind them.

Iodinated contrast (CT)

Contrast-associated AKI risk is meaningful mainly below eGFR 30, and more so with intra-arterial administration
Modern controlled evidence suggests the risk from IV contrast in stable outpatients has been substantially overstated historically
Mitigation is isotonic IV hydration; no benefit has been shown for N-acetylcysteine
Withhold metformin around the study if eGFR < 30, or in the presence of AKI

Gadolinium (MRI)

The concern is nephrogenic systemic fibrosis, not AKI
Risk is essentially confined to the older linear agents
With Group II macrocyclic agents — gadoterate meglumine, gadobutrol, gadoteridol — NSF risk is vanishingly low, including at eGFR < 30 and on dialysis
Routine eGFR screening before a Group II agent is no longer universally required
Non-contrast sequences remain an option — discuss with the reporting radiologist

Confirm which agent your imaging service uses, and what your local written screening protocol says, before quoting any threshold to a patient. Practice varies and the guidance has moved substantially.

Where cystatin C fits: for a frail, low-muscle-mass patient sitting near a decision threshold, a creatinine-based eGFR may be the reason a scan is either wrongly refused or wrongly waved through. It is a reasonable test to add before a contrast decision in that specific patient.

The other half of stagingA-stage · albuminuria

GFR is how much. Albuminuria is how well.

Everything so far has been about one axis. CKD staging has two, and the second one is the one we under-order. G-stage tells you the damage already accumulated. A-stage tells you what the glomerulus is doing right now — and it moves years earlier.

Why albumin in urine matters

The glomerulus is built to hold back large, negatively charged proteins. Albumin appearing in urine means the barrier is failing
Urine gives us an unusually direct, cheap, repeatable readout of glomerular endothelial function. Cardiology has vascular measures too — flow-mediated dilation, pulse-wave velocity — but none as easy to obtain in routine practice or as well validated against hard outcomes
Endothelial dysfunction precedes glomerulosclerosis by years, so albuminuria rises long before eGFR falls
G-stage and A-stage are independent risk factors for both cardiovascular and renal events

A patient with eGFR 75 and ACR 40 mg/mmol carries more risk than one with eGFR 55 and ACR 1.

The trap nobody mentions

The spot test we order is a ratio: urine albumin ÷ urine creatinine. It assumes everyone excretes about 1 g of creatinine per day.

Every criticism of serum creatinine applies to that denominator. A muscular patient excretes more urinary creatinine, so the ratio is falsely low. A cachectic patient excretes less, so the ratio is falsely high.

Same patient, same limitation, both ends of the workup. If the ACR doesn't fit the picture, a 24-hour urine albumin is the reference standard.

Transient albuminuria — not renal disease: vigorous exercise, prolonged standing (orthostatic proteinuria), fever, acute inflammation. Confirm on a repeat first-morning sample before labelling anyone.

Why A-stage changes management

ACE inhibitors / ARBs — the kidney-protective benefit beyond blood-pressure lowering appears to be confined to patients with albuminuria. Without it, they behave like any other antihypertensive
SGLT2 inhibitors — slow eGFR decline and reduce progression to end-stage disease and death (DAPA-CKD)
Finerenone — consider adding if albuminuria persists despite RAAS blockade and an SGLT2 inhibitor
Trend it. Unlike eGFR, albuminuria responds to treatment within months — it is a feedback loop, not just a label

Who to screen: everyone with diabetes (guideline-mandated), known CKD, hypertension, and anyone in whom you're already questioning the eGFR. Not universal screening — the false-positive rate is real.

Order-set implication: a renal assessment that returns an eGFR and no ACR is half a test. This is the easiest thing on the list for us to fix.

EscalationWhen estimating isn't good enough

When to stop estimating and measure

24-hour creatinine clearance

What: timed urine collection + serum creatinine.

Pros: patient-specific, no equation assumptions, available anywhere.

Cons: collection errors are common and usually invisible; still inflated by tubular secretion, typically overestimating GFR by 10–20%.

Whoever counsels the patient owns this result: the accuracy of the test is almost entirely determined by how well the collection is explained. Discard the first void, note the exact start time, collect everything for 24 h including the final void at the same clock time, keep refrigerated.

Measured GFR (mGFR)

What: plasma clearance of an exogenous filtration marker — iohexol, ⁵¹Cr-EDTA, ^99mTc-DTPA, or inulin historically.

Pros: the reference standard. Independent of muscle, steroids and inflammation.

Cons: costly, needs nuclear medicine or a specialist lab, several hours, timed sampling.

Reserve for: high-dose methotrexate, potential living kidney donors, carboplatin dosing in an extreme-body-habitus patient, and cases where eGFR_cr and eGFR_cys disagree profoundly.

One value vs a trend

A single eGFR is imprecise at the individual level — recall the 35–85 confidence interval around a reported 60.

Serial values on the same marker, same assay, same lab are far more informative than any one result, because the systematic error is largely constant within a patient. The direction of travel is the signal.

Don't re-stage a patient on one reading
Don't compare an outside lab's number to ours without checking equation and units
Do plot creatinine against weight over treatment cycles — that pairing is where the story shows up

And always pair the eGFR with an ACR. A renal workup with only one axis is half a test.

PracticeWhat this changes at the chairside

Your checklist

At booking and at the draw

Record an accurate current weight and height at every visit. Carboplatin dosing depends on it and BSA cannot be estimated by eye.
Ask about creatine or protein supplements and a large meat meal in the preceding 12 hours; note it on the request.
Note recent corticosteroid exposure if cystatin C is being sent.
Note any new trimethoprim, abemaciclib, ribociclib or similar agent started since the last panel.
Both markers go in a routine serum tube. No fasting required for either.

When results come back — flag to the doctor if

Creatinine has fallen while the patient is losing weight
Creatinine has risen >20% within two weeks of starting a listed drug, with a well patient
eGFR_cr and eGFR_cys differ by more than about 20 mL/min/1.73 m²
eGFR crosses 60, 45 or 30 in either direction
A carboplatin dose has been calculated from an indexed eGFR
A renal panel has come back without an ACR

The one sentence to take away: creatinine measures muscle as much as it measures kidney. When the muscle is changing — which in our patients it usually is — add cystatin C and use the combined equation.

Check yourselfFive questions

Five questions before we finish

1 · A 74-year-old man with metastatic gastric cancer has lost 12 kg. Creatinine 58 µmol/L, eGFR_cr 96. What is the most likely interpretation?

Low muscle mass means low creatinine production. The eGFR is falsely reassuring. Add cystatin C before any renally-dosed therapy.

2 · A patient on abemaciclib has a creatinine rise from 72 to 88 µmol/L at two weeks. She feels well, urine output normal. Best next step?

CDK4/6 inhibitors block tubular creatinine secretion. An unchanged cystatin C confirms GFR is intact and the drug can continue.

3 · Which of these raises cystatin C without any change in true GFR?

Corticosteroids increase cystatin C production. Meat and creatine affect creatinine, not cystatin C.

4 · The lab reports eGFR 60 mL/min/1.73 m² in a 45 kg, 150 cm woman (BSA ≈ 1.37 m²). What GFR should be used in the Calvert formula?

60 × (1.37 ÷ 1.73) ≈ 48 mL/min. Using 60 would overdose her carboplatin by roughly 20%.

5 · Which equation is most accurate against measured GFR when both markers are available?

The errors of the two markers are largely independent, so combining them cancels part of the bias. Prefer the combined equation whenever both values exist.

The five things worth remembering

Creatinine measures muscle as much as it measures kidney. Creatine, protein intake, body composition and immobility all move it without touching filtration.
A single eGFR is imprecise. A reported 60 carries a 95% confidence interval of roughly 35–85. Trends on the same assay beat single values.
Cystatin C fails on the opposite axis — steroids, thyroid disease, inflammation, adiposity. It is not a universal replacement.
The combined equation beats either marker alone. Use eGFR_cr-cys whenever both values exist.
Drug dosing needs de-indexed GFR in mL/min. The reported number is normalised to 1.73 m² and is the wrong input for the Calvert formula.

Sources & attributionFurther reading

Where this comes from

Primary inspiration

Core IM Podcast — "CKD Staging (Cr vs. Cystatin C, Albuminuria & more)," Mind the Gap. 8 November 2023. Blum C, Trivedi SP, Zhang Y, Katz G. Graphic by Ariella Coler-Reilly; audio by Yichi Zhang.
coreimpodcast.com/2023/11/08/ckd-staging-cr-vs-cystatin-c

Twenty-five minutes, CME-MOC available through ACP, and the clearest short treatment of this topic available for free. This deck began as an attempt to teach from that episode and its reference list, then expanded to cover supplement-driven creatinine elevation, drug dosing, and the interpretation of discordance. If you use one thing from this slide, use the podcast.

One point of departure. The episode's show notes and the widely circulated infographic describe cystatin C as neither secreted nor reabsorbed. Cystatin C is in fact almost completely reabsorbed by the proximal tubule and then catabolised there; it remains a valid serum marker because none returns to the circulation intact. A reader raised the same correction in the episode's comments. See slide 7.

Key primary literature

Inker LA, Eneanya ND, Coresh J, et al.	New creatinine- and cystatin C-based equations to estimate GFR without race. N Engl J Med 2021;385:1737–49. — the CKD-EPI 2021 equations implemented in the slide 13 calculator
Shafi T, Zhu X, Lirette ST, et al.	Quantifying individual-level inaccuracy in glomerular filtration rate estimation. Ann Intern Med 2022;175:1073–82. — the 35–85 confidence interval around a reported eGFR of 60
Gualano B, Ferreira DC, Sapienza MT, Seguro AC, Lancha AH Jr.	Effect of short-term high-dose creatine supplementation on measured GFR in a young man with a single kidney. Am J Kidney Dis 2010;55:e7–9. — the slide 2 dataset
de Souza e Silva A, Pertille A, Barbosa CGR, et al.	Is it time for a requiem for creatine supplementation-induced kidney failure? A narrative review. Nutrients 2023;15:1466.
Antonio J, Candow DG, Forbes SC, et al.	Common questions and misconceptions about creatine supplementation: what does the scientific evidence really show? J Int Soc Sports Nutr 2021;18:13.
Zhang X, Rule AD, McCulloch CE, Lieske JC, Ku E, Hsu CY.	Tubular secretion of creatinine and kidney function: an observational study. BMC Nephrol 2020;21:108.
Knight EL, Verhave JC, Spiegelman D, et al.	Factors influencing serum cystatin C levels other than renal function and the impact on renal function measurement. Kidney Int 2004;65:1416–21.
Ninomiya T, Perkovic V, de Galan BE, et al.	Albuminuria and kidney function independently predict cardiovascular and renal outcomes in diabetes. J Am Soc Nephrol 2009;20:1813–21.
Heerspink HJL, Stefánsson BV, Correa-Rotter R, et al.	Dapagliflozin in patients with chronic kidney disease (DAPA-CKD). N Engl J Med 2020;383:1436–46.
Calvert AH, Newell DR, Gumbrell LA, et al.	Carboplatin dosage: prospective evaluation of a simple formula based on renal function. J Clin Oncol 1989;7:1748–56.
KDIGO	2024 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease. Kidney Int 2024;105(4S):S117–S314.

Added at peer review — the newest evidence

These were surfaced when the draft was reviewed against the current literature, and several changed the content of the slides rather than merely supporting it.

Russel WA, Olde Engberink R, Creon A, et al.	Medications influencing serum cystatin C independent of measured GFR. J Am Soc Nephrol 2026; published online 17 Apr. doi:10.1681/ASN.0000001108. — iohexol-measured GFR in 5,595 adults; the basis for the revised confounder table on slide 9
Tsiaras A, Loufopoulos G, Theodoridis X, et al.	The effect of creatine supplementation on kidney function: a systematic review and meta-analysis of randomized controlled trials. J Ren Nutr 2026. doi:10.1053/j.jrn.2026.04.010
Kabiri Naeini E, Eskandari M, Mortazavi M, Gholaminejad A, Karevan N.	Effect of creatine supplementation on kidney function: a systematic review and meta-analysis. BMC Nephrol 2025;26:1–13. doi:10.1186/s12882-025-04558-6 — the pooled effect sizes quoted on slide 2
Stevens LA, Schmid CH, Greene T, et al.	Factors other than glomerular filtration rate affect serum cystatin C levels. Kidney Int 2009;75:652–60. — the ≈8.5% diabetes effect
Pottel H, Björk J, Rule AD, et al.	Cystatin C–based equation to estimate GFR without the inclusion of race and sex. N Engl J Med 2023;388:333–43.
Sandhu G, Adattini J, Armstrong Gordon E, et al.	Integrating international consensus guidelines for anticancer drug dosing in kidney dysfunction (ADDIKD) into everyday practice. EClinicalMedicine 2025. — the revised carboplatin guidance on slide 17
Tan DYZ, Wong BWX, Shen L, Li LJ, Yong EL.	Low creatinine to cystatin C ratio is associated with lower muscle volumes and poorer gait speeds in the longitudinal Integrated Women's Health Program cohort. Menopause 2025 Mar 18. doi:10.1097/GME.0000000000002524. — the slide 15 Singapore cohort. An erratum has been published; check it before quoting the CCR threshold
van Eeghen SJ, Wiepjes CM, den Heijer M, et al.	Cystatin C–based eGFR changes during gender-affirming hormone therapy. Clin J Am Soc Nephrol 2023;18(12):1545–54. — 260 transgender women, 285 transgender men; the marker-movement figures on slide 15
Yong EL, Wong BWX, Tan DYZ.	Beyond BMI: the Janus-like effect of muscle versus fat on midlife women's health. Ann Acad Med Singap 2025;54(2):125–8.
Department of Veterans Affairs / DoD	The Primary Care Management of Chronic Kidney Disease. Clinical Practice Guideline, 2025. — supports cystatin C where creatinine is unreliable

How this deck was made

Written and edited by Steven Tucker, M.D. (USA), FACP (USA), FAMS (Singapore) — Senior Consultant, Medical Oncology; Founder and Medical Director, Tucker Medical, Singapore — in collaboration with Claude Opus 5 (Anthropic), running with extended thinking enabled and live web retrieval. Every figure quoted in these slides was traced to its primary source during drafting rather than reproduced from model memory, and the CKD-EPI 2021 implementations behind the slide 13 calculator were independently computed and checked against published reference values.

The completed draft was then put through a slide-by-slide accuracy review using OpenEvidence. That review produced nine substantive changes, each verified against the primary source before being made: the magnitude of the creatine effect was recalibrated from case-report extremes to pooled randomised data; the cystatin C confounder table was rebuilt around iohexol-validated 2026 evidence, softening the thyroid claim and adding opioids and loop diuretics; the description of the cystatin C equation was corrected, since it does retain a sex coefficient; the carboplatin GFR cap was reframed as historical US guidance rather than current international consensus; and the claim made for albuminuria over cardiovascular measures was moderated. Anything the review flagged that could not be independently confirmed was left out.

Version 2.0 · 30 July 2026. Formatted to the Tucker Medical Brand Toolkit (TMED-017 v1.0); palette named from the ART-TM Colors swatch set, colour values sampled directly from the TuckerMedical_RGB_POS lockup.

VISIT	Novena Specialist Center, 8 Sinaran Drive, Suite 04-03, Singapore 307470
CALL	+65 6659 6918 · www.tuckermedical.com

Free to use, present and adapt for education with attribution. This is a teaching resource, not a clinical decision tool and not a substitute for clinical judgement. The carboplatin figures on slide 13 are illustrative only and must not be used for prescribing. Verify all thresholds, reference intervals and drug cut-offs against your own laboratory, formulary and current national guidance before applying them to a patient.

Two ways to measure a kidney

The most common cause of an "abnormal" creatinine we now see isn't kidney disease

Why the number moves

The cleanest single dataset

The eGFR isn't a number on a report. It's a gate.

Decisions gated by eGFR

What "eGFR" is really telling you

What a perfect filtration marker would look like

Creatinine

Where it comes from

How the kidney handles it

Why we still use it

Cystatin C

Where it comes from

How the kidney handles it

Practical notes

Side by side

What fools creatinine

What fools cystatin C

Pseudo-AKI: drugs that raise creatinine without touching the kidney

Common offenders

The pattern that identifies it

Which equation the lab is using

eGFRcr

eGFRcys

eGFRcr-cys

Legacy equations you'll still see

Two unit traps that change doses

Trap 1 — µmol/L vs mg/dL

CKD stage by eGFR

Trap 2 — indexed vs absolute GFR

See the discordance yourself

Absolute GFR (de-indexed)

Calvert carboplatin, AUC 5

Discordance is information, not noise

How to read the gap

Why the gap predicts outcomes

Menopause, MHT and the two markers

What the transition does

The Singapore evidence — IWHP

What MHT itself does to the markers

Does MHT help or harm the kidney? Honestly, unsettled

What to do in an MHT clinic

When to add cystatin C

Order it when…

Think twice when…

The creatine patient — two ways to settle it

Before you rely on it

Where a wrong GFR becomes a wrong dose

Carboplatin — the Calvert formula

Cisplatin

Other renally-sensitive agents

The steroid trap — and why some patients break both markers together

Falling muscle mass

Corticosteroids & inflammation

What to do

Who this describes — it is not a niche group

A worked example

Renal screening before imaging

Iodinated contrast (CT)

Gadolinium (MRI)

GFR is how much. Albuminuria is how well.

Why albumin in urine matters

The trap nobody mentions

Why A-stage changes management

When to stop estimating and measure

24-hour creatinine clearance

Measured GFR (mGFR)

One value vs a trend

Your checklist

At booking and at the draw

When results come back — flag to the doctor if

Five questions before we finish

The five things worth remembering

Where this comes from

Primary inspiration

Key primary literature

Added at peer review — the newest evidence

How this deck was made

eGFR_cr

eGFR_cys

eGFR_cr-cys