Case 05: The Salt Substitute

Commit to an answer before revealing the explanation. Use the section menu to navigate; each question and answer has its own link. Five-option evidence updates ask for the most defensible direction of change; discuss a justified adjacent response.

Session S2 · 2026-10-27 · in-session · 37 min Anchor: Chronic kidney disease — diabetic kidney disease and the four pillars · Integrates: chronic hyperkalemia on RAAS blockade; the expected creatinine bump; metformin, finerenone and GLP-1 RA by eGFR and potassium; sick-day rules and creatinine-bump triage; CKM framing, statin and nutrition counsel One-line hook: Every drug that protects a diabetic kidney nudges creatinine up and potassium up. A new PA has to tell apart the bump that is the medicine working, the bump that is the neighbor's advice, and the day that is an ED visit.

Learning Objectives

Answers for this section: Q1 · Q2 · Q3 · Q4 · Q5 · Q6 · Q7 · Q8 · Q9 · Q10 · Q11

1. Recognize typical diabetic kidney disease (long diabetes duration, retinopathy, albuminuria with a bland sediment, slow eGFR decline) and name the findings that should make you look for another cause.
2. Start the first two pillars (ARB and SGLT2 inhibitor), state the potassium (5.0 mEq/L or lower; extra monitoring at 4.8–5.0) and eGFR (25 or more) entry rules for finerenone, know that a GLP-1 receptor agonist does not wait for the potassium, and adjust metformin by eGFR.
3. Interpret a creatinine rise after an ARB or SGLT2 inhibitor: accept a rise of 30 % or less, act on a rise above 30 % or a potassium of 5.5 mEq/L or more.
4. Work up chronic hyperkalemia on RAAS blockade (salt substitute, NSAID, diet, trimethoprim) and apply the contributor-removal steps before reaching for a binder or stopping the ARB.
5. Give a sick-day plan (SADMANS) with an individualized restart plan, and triage an outpatient creatinine bump to same-day ED versus a 48–72-hour recheck.

Warm-Up (individual commit before the case)

Answers for this section: Q1 · Q2 · Q3

Part 1 — Presentation

Answers for this section: Q4 · Q5

Day 0 — primary care clinic, scheduled visit. Mr. Dale Hinrichs is a 64-year-old retired rural mail carrier who comes in with his wife for his diabetes follow-up. He feels fine. He has had type 2 diabetes for 12 years and hypertension for 10. Last year's eye exam found mild non-proliferative ("background") diabetic retinopathy — small retinal hemorrhages and microaneurysms, no new vessels, no treatment needed yet. He has gout, two flares in the past three years, the last one eight months ago. He has never smoked and drinks two beers on a weekend. His father had diabetes and "died on dialysis." He has no urinary symptoms, no edema, no foamy urine that he has noticed, and he has never seen blood in his urine.

His creatinine has drifted up slowly: 1.7 mg/dL two years ago (eGFR 45), 1.8 fourteen months ago (eGFR 42). Nobody has yet called it chronic kidney disease out loud.

Vitals: BP 138/84 mmHg seated after 5 minutes of rest, repeated 136/84 · pulse 76 regular · afebrile · weight 96 kg · height 176 cm · BMI 31.0 kg/m².

Exam: Comfortable. No jugular venous distension. Lungs clear. Heart regular, no murmur. Abdomen soft, no bruit. No leg edema. Diminished vibration sense at both great toes; monofilament intact. Feet without ulcers.

No over-the-counter products or supplements at this visit. No potassium supplement. No NSAID "except for a gout flare, and that was months ago."

Today's labs (fasting, drawn before the visit; all values steady state):

Urinalysis (dipstick + microscopy): specific gravity 1.015 · pH 5.5 · protein 2+ · glucose trace · blood negative · leukocyte esterase negative · nitrite negative. Microscopy: 0–2 RBC/hpf, 0–2 WBC/hpf, no casts, no crystals. A bland sediment with proteinuria.

Renal ultrasound (14 months ago, ordered for the first creatinine of 1.8): right kidney 11.2 cm, left 11.5 cm, mildly increased cortical echogenicity, no hydronephrosis, no stones, no masses.

A word on the arithmetic before the questions. The eGFR of 39 is what CKD-EPI 2021 gives for a creatinine of 1.9 in a 64-year-old man. The slope is 45 → 42 → 39 over two years, about 3 mL/min per year — a slow decline, not a rapid one (rapid is more than 5 mL/min per year). The BUN:creatinine ratio of 14:1 is what CKD looks like at steady state. The anion gap is 12, so there is no hidden acid-base story. Hemoglobin 13.1 is a whisker under the line — iron studies at the next visit; CKD anemia is the Tired Teacher's case (Case 04), not this one.

Pause — commit before the reveal

What you tell him before he leaves. Mr. Hinrichs asks why he takes a cholesterol pill when "my cholesterol is fine." The honest answer is the CKM frame — cardiovascular-kidney-metabolic syndrome, the 2026 guideline's name for the fact that his blood pressure, his glucose, his lipids and his kidney are one disease with shared plumbing. Type 2 diabetes plus hypertension is CKM stage 2; very-high-risk CKD (his red cell on the heat map) is a risk equivalent that moves him to stage 3. Stage 3 means he is treated as if he already had cardiovascular disease: the statin continues whatever the LDL says, the BP target is below 130/80 by the 2025 AHA/ACC guideline (KDIGO 2021 suggests a standardized systolic below 120 when tolerated — same patient, different measurement method), and the kidney drugs are also heart drugs. One problem, four dials.

He also gets three practical instructions: cut sodium to under 2 g a day; expect the empagliflozin to put sugar in his urine and to pull his kidney number down a little at first ("that is the medicine lowering the pressure inside the filter, not damage"); and a written sick-day card — if he is vomiting, has diarrhea, or cannot eat or drink for a day, he stops the empagliflozin, the losartan and the metformin and calls. The card goes in his wallet. Hold that thought.

Part 2 — Data

Answers for this section: Q6 · Q7 · Q8

Day 14 — the 2-week lab check. He feels well. He has been "really good about the salt": his neighbor, whose wife is a nurse, told him that the way to cut salt is to switch to NoSalt, so he has been using it on everything for 12 days — "the kids love it on the popcorn." Nine days ago his left big toe flared. He took what he always takes for gout, naproxen sodium 220 mg, two tablets twice a day, and it has mostly settled; he is still taking it "to be safe." Neither item is in his chart. He has not been sick, is eating and drinking normally, and has not missed a dose of anything.

Vitals: BP 128/78 mmHg · pulse 74 · weight 94.5 kg (down 1.5 kg).

Urinalysis (Day 14): specific gravity 1.012 · protein 2+ · glucose 3+ · blood negative. Microscopy bland.

ECG (Day 14): sinus rhythm 74/min, PR 160 ms, QRS 88 ms, normal T waves, no peaking. Read side by side with the Day 0 tracing — unchanged.

Three notes on the numbers. The 3+ glucose in the urine is the empagliflozin doing its job, not new diabetes. The lab prints an eGFR of 33 because that is what the formula returns for a creatinine of 2.2; the number is real, but it is the hemodynamic adjustment, not six lost mL/min of nephrons. And the bicarbonate slipped from 23 to 21 with the potassium rising: that pairing — mild normal-gap acidosis plus hyperkalemia in a diabetic on RAAS blockade and an NSAID — is the type 4 renal tubular acidosis pattern (too little aldosterone effect at the collecting duct). The naproxen he just added makes it worse by suppressing renin; the salt substitute simply piles potassium onto a collecting duct that cannot clear it.

Pause — commit before the reveal

The gout. Naproxen is out for good — any NSAID in a man with eGFR in the 30s on an ARB is the "triple whammy" waiting for its third partner (a diuretic or a stomach bug). For the next flare he gets colchicine at a dose reduced for his eGFR or a short prednisone taper, and the clinic plans urate-lowering therapy once the acute problem settles. Two small gifts: losartan is mildly uricosuric (it helps the kidney excrete uric acid), and SGLT2 inhibitors lower uric acid a little too.

Part 3 — Course and Decisions

Answers for this section: Q9 · Q10 · Q11

Day 28 — four weeks on empagliflozin, two weeks off NoSalt and naproxen. The pre-visit BMP is back.

Day 28, in full: potassium 4.9 mEq/L, creatinine 2.0 mg/dL (eGFR 37), bicarbonate 23; BP 126/76. Semaglutide 0.25 mg subcutaneously once weekly starts today for weight and cardiovascular risk, titrated at 4-week steps; finerenone is penciled in for the next visit if the potassium is 4.8 or lower on the pre-visit draw. He is reminded of the sick-day card.

Month 3 — a phone call, then a same-day slot. His wife calls. Their grandson brought a stomach bug home; Mr. Hinrichs has had vomiting and watery diarrhea for three days, has kept down "a little broth and some crackers," and feels dizzy when he stands. He has taken all of his pills every day, "because the doctor said they protect my kidneys." He has not urinated since early morning. Nobody looked at the sick-day card.

Vitals in the clinic: seated BP 112/68 mmHg, pulse 94 · standing at 1 minute BP 92/58, pulse 112, lightheaded · temperature 37.6 °C · weight 93 kg (down 3 kg from Day 28). Dry mucous membranes, no jugular venous distension, lungs clear, abdomen soft and mildly tender diffusely, no edema.

Urinalysis: specific gravity 1.024 · glucose 3+ · protein 2+ · blood negative · ketones trace. Microscopy: hyaline casts, otherwise bland. ECG: sinus tachycardia 108/min, normal intervals, no T-wave peaking.

Read it in order. The creatinine is up 0.9 mg/dL (45 %) in days — AKI by the KDIGO criterion of a rise of 0.3 mg/dL or more within 48 hours. The other criterion, 1.5 times baseline (3.0 mg/dL here), is one more day of diarrhea away. The BUN rose faster than the creatinine, so the ratio climbed from 13:1 to 20:1 — the prerenal pattern, with nothing else (no steroids, no GI bleed, no protein load) to explain the urea. The urine is concentrated and bland with hyaline casts, which fits an under-perfused but intact kidney. The potassium is up because less filtration means less sodium and less flow reaching the collecting duct, so less potassium is secreted — and the ARB is still blocking what aldosterone is left. The bicarbonate is down because diarrhea loses bicarbonate and because a poorly perfused kidney retains acid. The anion gap stays 12 — this is not ketoacidosis, which matters on an SGLT2 inhibitor: euglycemic diabetic ketoacidosis presents with exactly this story and a near-normal glucose, so the trace ketones earn a venous blood gas and a serum beta-hydroxybutyrate in the ED.

Pause — commit before the reveal

What he should have done, and what the card said. SADMANS: Sulfonylureas, ACE inhibitors, Diuretics, Metformin, ARBs (and ARNI), NSAIDs, SGLT2 inhibitors — the drugs to pause during vomiting, diarrhea, fever or a day without eating and drinking; the mineralocorticoid antagonists (spironolactone, eplerenone, finerenone) are held too. Each has a mechanism: the SGLT2 inhibitor continues to dump sodium and glucose into the urine when he cannot replace them and adds ketoacidosis risk; the ARB blocks the angiotensin his kidney needs to hold filtration when volume is low; metformin accumulates as clearance falls; an NSAID takes away the afferent dilation that protects the glomerulus when perfusion is poor. The GLP-1 receptor agonist is not on the mnemonic, but a weekly injection that already makes him queasy is worth discussing before the next dose. What he should not stop: his statin, and — had he been on one — a beta-blocker. The restart rule is Q11.

Resolution and Teaching Points

Answers for this section: Q1 · Q2 · Q3 · Q4 · Q5 · Q6 · Q7 · Q8 · Q9 · Q10 · Q11

In the ED he received 2 L of isotonic crystalloid over four hours; the potassium fell to 5.2 mEq/L with fluid alone, the venous gas and beta-hydroxybutyrate were normal, and he was discharged that night with every SADMANS drug on hold and a BMP at 48 hours (creatinine 2.5, potassium 5.0). At one week — creatinine 2.1, potassium 4.8, eating normally — losartan 100 mg was restarted. At two weeks empagliflozin came back. At three weeks the creatinine was 2.0 mg/dL (eGFR 37) on two draws, and metformin returned at 500 mg twice daily. Semaglutide resumed at 0.25 mg and was titrated to 0.5 mg a month later.

The nephrology visit that was requested on Day 0 landed in month 4. Three things happened there that a new PA should be able to do in primary care. First, the kidney failure risk equation (KFRE) was run — a calculator that uses age, sex, eGFR and UACR to give two numbers: the chance of needing dialysis or a transplant within 2 years and within 5 years. Run with his steady-state eGFR of 37 and his UACR of 900 mg/g, his 4-variable result was about 6 % at 2 years and about 18 % at 5 years (North American calibration of the 4-variable equation — the calculator has two calibrations; use the North American one here). KDIGO uses a 5-year risk of 3–5 % as the nephrology-referral threshold, a 2-year risk above 10 % for multidisciplinary kidney care, and a 2-year risk above 40 % to start dialysis education and access planning. He clears the referral line comfortably — and he already qualified on the heat map alone (G3b with A3) and would also have qualified for a decline of more than 5 mL/min a year had he had one. The referral is co-management, not a hand-off. Second, the dietitian gave the counsel in plain numbers: sodium under 2 g (2,000 mg) a day without a potassium salt substitute; potassium individualized — no blanket ban on fruit and vegetables now that the level is 4.8, but no NoSalt, no "lite" salt, and labels read on reduced-sodium broths; protein about 0.8 g per kilogram a day — about 75 g on his actual weight, closer to 65 g if the dietitian uses an adjusted weight for his BMI — reassessed as the semaglutide takes weight off (a very-low-protein diet is specialist territory, not a goal); and phosphorus additives ("phos-" on an ingredient list) avoided before natural phosphorus is restricted. Third, with the potassium 4.7 on two checks, finerenone 10 mg daily was started with a potassium check at 4 weeks.

His gout has not flared since the naproxen stopped; colchicine at a reduced dose is in the cabinet for the next one, and allopurinol was started low and titrated. A new sick-day card is in his wallet. His wife has one in her purse.

Clinical Pearl — The 30 % rule has two halves The >30% creatinine threshold belongs to reassessment after starting or increasing ACEi/ARB (KDIGO 2024, 3.6.4); it is not proof that every smaller rise is harmless. SGLT2-related initial eGFR dips are commonly reversible. Review volume, NSAIDs, potassium and serial values before changing protective therapy.

Clinical Pearl — Hyperkalemia on RAAS blockade is a work-up, not a reflex Before touching the ARB, hunt for the reversible contributors: potassium-chloride salt substitutes and "lite" salt, NSAIDs, trimethoprim, potassium supplements, potassium-sparing diuretics, and uncontrolled glucose. Then the ladder: diet, loop diuretic if volume-expanded, binder (patiromer or sodium zirconium cyclosilicate). Stopping the ARB is the bottom rung.

Clinical Pearl — Metformin has three bands and one hold rule Do not start below eGFR 45; halve it (maximum 1,000 mg a day) between 30 and 45; stop below 30. Separately, hold it during any illness that threatens filtration — vomiting, diarrhea, dehydration, IV iodinated contrast with AKI or eGFR <30 under the ACR approach; the FDA label and some local policies are stricter — whatever the baseline band.

Clinical Pearl — The four pillars are sequenced by potassium and indication, not by enthusiasm ARB (or ACE inhibitor) and SGLT2 inhibitor first, together or within weeks. Finerenone only when potassium is 5.0 or lower (extra monitoring at 4.8–5.0) and eGFR 25 or more. A GLP-1 receptor agonist whenever glucose, weight or cardiovascular risk calls for it — it does not wait for the potassium.

Clinical Pearl — Triage a creatinine bump on four things Triage by potassium and its trajectory, ECG, symptoms, volume, intake, urine output and reliable follow-up. Stable mild hyperkalemia may be managed with prompt outpatient remeasurement; acute illness, oliguria, hypotension or a rising potassium requires same-day assessment. IV fluid is appropriate for depletion, not for every creatinine rise.

Pitfall — Stopping the pillar for the dip Avoid reflex withdrawal for a modest stable hemodynamic dip, but do not call every creatinine rise protective. Review perfusion, concurrent illness, nephrotoxins, potassium and repeat values. Continue, reduce or temporarily hold treatment according to that assessment.

Pitfall — A sick-day card with no restart line Half the harm of sick-day rules comes from patients who stop the ARB during a stomach bug and never start it again. Every hold comes with a restart trigger (eating and drinking 24–48 hours, BMP checked) and an order (ARB, then SGLT2 inhibitor, then metformin at the eGFR-appropriate dose), written down and followed up with a lab.

Concept Links

Answers for this section: Q1 · Q2 · Q3 · Q4 · Q5 · Q6 · Q7 · Q8 · Q9 · Q10 · Q11

Assigned reading: the explanations in this case and the question-linked course evidence guide. The objective map distinguishes tested decisions from concepts taught in explanations.

Concepts: aki-14, aki-15, aki-27, ckd-02, ckd-06, ckd-11, ckd-12, ckd-13, ckd-17, ckd-20, htn-cardiorenal-07, htn-cardiorenal-09, htn-cardiorenal-24, htn-cardiorenal-25, lytes-14, structural-23, structural-24.