Clinical Cases in Pharmacology Clinical Cases  ·  Pulmonary Vol. I  ·  Sleep Medicine, Neuromuscular, and Skeletal  ·  The Device the Referral Almost Ordered
Pulmonary Vol. I, Case 0003 — Sleep Medicine, Neuromuscular, and Skeletal

The Device the Referral Almost Ordered

A sleep study recommends adaptive servo-ventilation for Cheyne-Stokes breathing before anyone checks it against a well-established mortality signal in exactly this population.

Abbreviations, terms, and other agents mentioned in this case CSA — central sleep apnea  ·  ASV — adaptive servo-ventilation  ·  HFrEF — heart failure with reduced ejection fraction  ·  EF — ejection fraction  ·  GDMT — guideline-directed medical therapy  ·  ARNI — angiotensin receptor-neprilysin inhibitor  ·  MRA — mineralocorticoid receptor antagonist  ·  SGLT2 — sodium-glucose cotransporter-2  ·  eGFR — estimated glomerular filtration rate
Presentation

P.K. was diagnosed with heart failure four months ago after a routine physical turned up an ejection fraction of 30%, a number that startled him more than any symptom had — he'd chalked his fatigue up to being 63 and out of shape, not a heart working at less than a third of what it should. A sleep study, ordered to work up that same fatigue, found a breathing pattern his sleep physician recognized immediately: Cheyne-Stokes respiration, a crescendo-decrescendo waxing and waning of breath depth with central apneic pauses at the trough, present through most of his overnight recording. The study's own automated report recommended adaptive servo-ventilation, and that recommendation had already been forwarded toward a device order before his new cardiologist, reviewing the referral, stopped it.

The reason is not a matter of clinical preference. SERVE-HF, a large randomized trial published a decade ago, found that ASV increased cardiovascular mortality in patients with HFrEF and an ejection fraction at or below 45% with predominant central sleep apnea — P.K.'s exact population — and the finding was strong and consistent enough that ASV became, and remains, contraindicated in this specific group across every major guideline that has addressed it since. It is not a soft caution or a relative preference; it is a therapy genuinely capable of harming him, recommended by a sleep study algorithm built before that finding existed or applied broadly enough, and still generated because most automated scoring software has no field for "diagnosis-specific contraindication" — it reads an AHI and proposes the device that lowers it, without knowing his ejection fraction at all. What the team actually has in front of them, once ASV is off the table entirely, is a real, still-open question: his carvedilol, sacubitril-valsartan, spironolactone, and dapagliflozin are each below target dose, started at diagnosis but never fully uptitrated at his prior clinic, and Cheyne-Stokes respiration in HFrEF often improves substantially as ejection fraction recovers on optimized therapy — but "often" is not "always," and he has real fatigue now, not in the months uptitration will take.

P.K. · 63 Sleep Medicine / Cardiology Joint Review
History
HFrEF diagnosed 4 months ago, EF 30% at diagnosis, no prior cardiac history
Sleep study finding
Cheyne-Stokes respiration, predominant central sleep apnea, AHI 34
Automated report recommendation
ASV — contraindicated in his population per SERVE-HF, not yet ordered
Current HFrEF therapy
Carvedilol, sacubitril-valsartan, spironolactone, dapagliflozin — all below target dose
EF, most recent
30%, unchanged since diagnosis
Symptoms
Persistent fatigue, mild dyspnea on exertion
Renal function
eGFR 68, adequate for continued GDMT uptitration
Electrolytes
Potassium 4.2, normal

Joint sleep medicine / cardiology review

Pulmonologist Opening

The automated report recommended ASV, and I want to be direct about why that recommendation does not go forward: SERVE-HF found increased cardiovascular mortality with ASV in patients with EF at or below 45% and predominant central sleep apnea — his exact profile. This isn't a relative preference or a softer caution. It's contraindicated. Before we talk about anything else, I want it on record that ASV is off the table entirely, not just deprioritized.

Cardiologist Response

Agreed, fully, and thank you for catching that before it reached an order. Once ASV is off the table, I'd point to something else in his chart: none of his four heart failure medications are at target dose yet. Cheyne-Stokes respiration in HFrEF is frequently a downstream effect of the underlying cardiac physiology, and it often improves substantially as ejection fraction recovers on fully optimized therapy. I'd rather maximize what we know helps the underlying disease before reaching for anything adjunct.

Clinical Pharmacologist Final

I don't disagree that GDMT optimization is the real, primary strategy here. But uptitration realistically takes months, and his fatigue is happening now, not in month three. Acetazolamide carries a conditional recommendation for heart-failure-related CSA in the 2025 AASM guideline — low certainty, but a real one — for reducing central apnea events through carbonic anhydrase inhibition stimulating ventilatory drive — genuinely worth naming as a bridge option, even if neither of you wants to start it today.

Waiting for "fully optimized therapy" before even discussing an adjunct treats titration as a single endpoint rather than a months-long process he has to live through — I'd rather we agree on the threshold for adding acetazolamide now, rather than only revisiting it once titration is already complete.

Regimen selected
Carvedilol
Beta-Blocker · Uptitration toward target dose begun today
One of four foundational HFrEF agents left below target dose at his prior clinic; escalation begun as the primary strategy for his central sleep apnea.
Sacubitril-Valsartan
ARNI · Uptitration toward target dose begun today
Second of the four GDMT agents being escalated in parallel, given the demonstrated relationship between EF recovery and Cheyne-Stokes improvement.
Spironolactone
MRA · Uptitration toward target dose begun today, with potassium monitoring
Third GDMT agent escalated; potassium and renal function to be rechecked given his currently normal but not yet stress-tested electrolyte panel.
Dapagliflozin
SGLT2 Inhibitor · Uptitration toward target dose begun today
Fourth GDMT agent escalated, completing the quadruple-therapy platform the group is relying on as the primary strategy.
Adaptive Servo-Ventilation — Contraindicated, Ruled Out
Device Therapy · Not adopted, safety contraindication
Removed entirely from consideration given SERVE-HF's demonstrated mortality signal in his exact population (HFrEF, EF ≤45%, predominant CSA).
Acetazolamide — Held in Reserve
Carbonic Anhydrase Inhibitor · Contingent, not started today
Named explicitly as the next step if his symptoms remain significant partway through GDMT uptitration, rather than only reconsidered once titration is fully complete.
Where this was left

Agreed: ASV formally removed from the plan; GDMT uptitration begins today with a structured dose-escalation schedule and a repeat sleep study planned once target doses are reached, roughly three months out.

Not fully agreed: the pharmacologist would prefer a pre-set symptom threshold at which acetazolamide starts before GDMT titration completes, rather than waiting for the full three months; the cardiologist would rather see the uptitration through first, given acetazolamide's own real side-effect burden (paresthesias, metabolic acidosis, renal stone risk) and the strength of the GDMT-improvement evidence. The two agreed to revisit the threshold question at the one-month follow-up rather than resolve it today.

Educational content only — a composite teaching case, not a real patient encounter or a substitute for clinical guidance. About These Cases →