A Fontan Patient in Septic Shock: When a Fluid Bolus Is the Wrong Reflex
A 27-year-old woman with hypoplastic left heart syndrome palliated with an extracardiac Fontan conduit at age 3 presents at 2 a.m. with 36 hours of fever, vomiting and new confusion. Her baseline saturation is 94% breathing room air; tonight she is at 86% on 4 L. Blood pressure is 84/52 mmHg, heart rate 128/min, temperature 39.1 °C. She has already received 1 L of crystalloid, after which her blood pressure rose briefly and her saturation fell to 82%. Lactate 4.6 mmol/L, creatinine 1.6 mg/dL (baseline 0.9). Bedside ultrasound shows a dilated, plethoric inferior vena cava, a systemic ventricular ejection fraction of 45% (unchanged from last year) and no pericardial effusion. Blood cultures are drawn and broad-spectrum antibiotics are being prepared.
1. Why did the fluid bolus lower her oxygen saturation, and how should that change your resuscitation?
Read the worked answer
Fontan flow is pressure-driven, not pumped. Pulmonary blood flow approximates (systemic venous pressure − atrial pressure) ÷ pulmonary vascular resistance. There is no subpulmonary ventricle to recruit when vascular tone falls, so sepsis behaves physiologically like an obstructive lesion.
- Septic vasodilation lowers systemic vascular resistance; venous return and the upstream pressure that drives transpulmonary flow fall together. Desaturation can therefore precede or accompany hypotension - her falling saturation is a hemodynamic signal, not just pulmonary congestion.
- Crystalloid raises venous pressure, but with a non-compliant systemic ventricle the increment is transmitted to the pulmonary venous atrium. The transpulmonary gradient barely changes while pulmonary edema and pleural effusion develop. The transient blood-pressure rise followed by desaturation is the classic signature of this error.
Implication: the therapeutic lever is vascular tone, not volume. Start norepinephrine early and titrate to mean pressure. Small aliquots (5–10 mL/kg) with immediate reassessment are defensible because of vomiting, but large boluses are harmful. Antibiotics within the hour and source control are equally urgent.
2. Outline your first two hours, including which studies you order and why.
Read the worked answer
- Setting and monitoring: intensive care, arterial line, central access, hourly urine output, continuous oximetry. Accept saturations in the high 80s if perfusion and lactate improve.
- Vasopressor first: norepinephrine to a mean pressure of about 65 mmHg, escalating as needed. Consider stress-dose hydrocortisone if refractory.
- Antibiotics within one hour, fluid aliquots only with reassessment after each.
- Laboratory work: blood gas, serial lactate, coagulation studies, hepatic panel, creatinine, blood cultures (already drawn), and later stool alpha-1 antitrypsin if there is diarrhoea or hypoalbuminemia.
- Imaging: bedside echocardiography to assess filling, conduit flow, atrioventricular valve regurgitation and effusion - but a normal transthoracic study does not exclude conduit thrombus.
Reasoning: in a Fontan, obstruction at conduit, pulmonary artery, pulmonary vein or atrioventricular valve level mimics sepsis-related low output and will not respond to vasopressors. Before attributing everything to infection, prove that the circuit is patent - contrast computed tomography or catheterization if the response to tone support is poor.
3. At hour 4, on norepinephrine 0.2 mcg/kg/min, mean pressure is 58 mmHg, lactate 5.6 mmol/L, saturation 80%, central venous pressure 22 mmHg, urine 5 mL/hour. Another litre is proposed. What is your judgement?
Read the worked answer
Refuse the litre. A central venous pressure of 22 mmHg with falling output and saturations despite high-dose vasopressor is not the picture of simple vasoplegia, and additional volume will raise pulmonary venous pressure without improving transpulmonary flow.
Reasons to pivot toward a mechanical or pump-failure explanation:
- Obstruction anywhere in the Fontan pathway produces exactly this pattern - rising venous pressure, falling saturation, no pressor response.
- The systemic ventricle may be failing in the setting of Sepsis, fever and tachycardia; ejection fraction on a single bedside study underestimates this.
Actions: repeat focused ultrasound including hepatic vein Doppler (loss of phasic flow suggests obstruction), then urgent contrast computed tomography or catheterization. Call congenital cardiac surgery and the interventional team. If thrombus is confirmed, start systemic heparin and discuss catheter-directed therapy; if the ventricle is failing, add inotrope and consider epicardial pacing or arrhythmia control. Re-examine for a new atrioventricular valve problem or effusion. Document the fluid decision explicitly, because nursing and resident teams will otherwise repeat it.