Case October 2026 - Dexamethasone-Associated Hypertrophy And Ductal Closure

Clinical context

Case of an infant born extremely preterm at 24 weeks' gestation. The patient required intubation and surfactant at birth. A first attempt at extubation failed, and she was reintubated and escalated to high-frequency jet ventilation with significant respiratory support. In the setting of evolving ventilator dependence, a prolonged dexamethasone course was started on day of life 9 (corrected age 25+2 weeks). The course was based on the DART protocol, with each step prolonged (doubled). A targeted neonatal echocardiogram (TNE) was obtained at the initiation of dexamethasone. It showed a large, hemodynamically significant patent ductus arteriosus (hsPDA) with the classic phenotype of pulmonary overcirculation and systemic steal (details below) - see pre-dexamethasone TNE.

On day 10 of dexamethasone (day of life 18, corrected age 26+4 weeks, weight 640 g), the infant had improved considerably. She had been successfully extubated to non-invasive ventilation (NIMV 14/8, FiO₂ 23%). Blood pressure was 64/34 mmHg without any inotropic support. She was on full enteral feeds with no central lines. The team requested a follow-up TNE with two questions: had the ductus closed, and was there myocardial wall thickening in the context of dexamethasone exposure?

The study was limited by extensive pulmonary artefacts. It nevertheless showed a closed ductus, biventricular hypertrophy with a posterior LV wall z-score of +5, a narrowed LV cavity, intracavitary flow acceleration and systolic notching of the LVOT Doppler envelope. 

Baseline TNE: a volume-loaded left heart

The first TNE, obtained at the start of dexamethasone, showed a large hsPDA with a predominant left-to-right shunt and a full set of markers of hemodynamic significance:

• Pulmonary overcirculation: dilated left atrium and left ventricle, with a high D-wave on pulmonary venous Doppler. Increased ductal shunt flow raises pulmonary blood flow, and therefore pulmonary venous return to the LA. In the preterm infant, pulmonary venous flow is often near-continuous, and a high diastolic (D) velocity reflects increased flow volume and LA loading.

• Large ductal diameter the size of the left pulmonary artery with left to right pulsatile doppler pattern (unrestrictive pattern). Low diastolic velocity on the ductal Doppler and positive diastolic velocities on the left pulmonary artery and right pulmonary artery Dopplers.

• Systemic steal: holodiastolic retrograde flow in the descending aorta, and absent end-diastolic flow in the celiac artery. During diastole, blood runs off from the aorta into the lower-resistance pulmonary circuit, at the expense of mesenteric perfusion.

At this point, the LV was operating under high preload and high total output. As we discussed in the May 2026 case (post-ligation syndrome), a large ductus also partially protects the LV from afterload, because part of the systemic arterial tree is decompressed into the pulmonary vascular bed. A dilated, volume-loaded LV cavity can also hide early wall thickening: a thick wall is less obvious when the cavity is large and stented by significant preload.

What changed between the two studies?

In ten days, three things happened at the same time.

1. The ductus closed. Several factors likely contributed. Postnatal corticosteroids have been associated with ductal constriction (reference, reference 2), possibly by reducing ductal sensitivity to prostaglandins and increasing its sensitivity to oxygen. Better lung aeration, lower mean airway pressure after extubation and advancing postnatal age may also have played a role. Once the ductus closed, pulmonary venous return fell and the LA and LV were no longer volume-loaded. The LV cavity became smaller (decreased preload).

2. The LV lost its ductal "escape route". Systemic afterload rose as the low-resistance ductal runoff disappeared. Dexamethasone can add to this, because it raises systemic vascular resistance and blood pressure. This can contribute to wall stress and to the reactive myocardial hypertrophy.

3. The myocardium thickened. Dexamethasone is a potent trigger of cardiomyocyte hypertrophy (see below). Over the same ten days, the posterior LV wall reached a z-score of +5, with septal and subjective RV hypertrophy.

Put together, the LV went from a large, volume-loaded, afterload-protected chamber to a small, thick-walled, hyperdynamic chamber ejecting against a higher afterload. This is the substrate for dynamic intracavitary obstruction.

TNE on day 10 of dexamethasone

Shunts

• PDA closed.

• Patent foramen ovale with left-to-right shunting.

• No retrograde flow in the pulmonary veins. Normal forward flow in the IVC and SVC.

Hypertrophy

• Subjective biventricular hypertrophy.

• LV posterior wall thickness in end-diastole: 0.38 cm (z-score +5).

• Interventricular septum thickness in end-diastole: 0.29 cm (z-score +2.9).

• Relatively narrowed LV cavity.

Left ventricle

• Preserved systolic function: EF 61% (apical 4-chamber) and SF 37%.

• Aortic VTI 9.0 cm, LVOT diameter 0.43 cm, heart rate 150 bpm: LV output 306 mL/kg/min (upper range of normal; with the ductus closed, this reflects systemic blood flow).

• Intracavitary flow acceleration within the LV, consistent with dynamic acceleration in a hypertrophied cavity.

• Systolic notching on the LVOT pulsed-wave Doppler, with a "dagger-shaped" envelope.

• No LVOT obstruction on 2D or color Doppler, and no systolic anterior motion (SAM) of the mitral valve.

• Mitral E/A 0.54.

• Trivial mitral regurgitation.

Right ventricle and pulmonary hemodynamics

• Normal RV systolic function: TAPSE 0.55 cm (normal for age).

• Pulmonary VTI 11.2 cm.

• Tricuspid E/A 0.54, with subjective RV hypertrophy.

• PAAT/RVET 0.53 and round septum at end-systole: no signs of elevated pulmonary pressure.

• Trivial tricuspid regurgitation.

Impression: closed ductus, preserved biventricular systolic function, and dexamethasone-associated biventricular hypertrophy with dynamic LV intracavitary and outflow flow acceleration, without fixed LVOT obstruction. The reversed E/A patterns suggest some impaired relaxation (restrictive filling physiology) in both ventricles.

Why do corticosteroids are associated with myocardial hypertrophy?

Hypertrophic cardiomyopathy in preterm infants treated with dexamethasone for chronic lung disease was first described in the early 1990s. Prospective echocardiographic studies later showed that septal and posterior wall thickening can appear within days of starting treatment, even with short courses. Several mechanisms probably act together:

• Direct effect on the myocyte: glucocorticoid receptors are expressed in the fetal and neonatal heart. Their activation promotes myocyte maturation and protein synthesis, which increases myocyte size and wall thickness.

• Increased afterload: dexamethasone raises blood pressure through greater vascular sensitivity to catecholamines and angiotensin II. The LV responds with concentric remodelling.

• Metabolic effects: steroid-induced hyperglycemia and hyperinsulinemia promote myocardial growth, similar to the hypertrophic cardiomyopathy seen in infants of diabetic mothers.

• Loss of ductal decompression: in this infant, ductal closure added a sudden rise in afterload and a fall in preload during the same period.

The hypertrophy is usually transient. Wall thickness generally regresses over weeks after the steroid is weaned or stopped. The effect seems related to dose and duration, which matters for prolonged or higher-dose regimens like the one used here. Hydrocortisone appears to carry less risk than dexamethasone, although data are more limited.

In most infants, the hypertrophy is clinically silent. The concern is the minority in whom a small, thick-walled, hyperdynamic LV develops dynamic obstruction, reduced stroke volume and impaired diastolic filling. This risk rises sharply when tachycardia, hypovolemia or inotropes are added. For more information, see this review by our group: "Plessas-Azurduy, P.; Lapointe, A.; Wutthigate, P.; Spénard, S.; Villeneuve, A.; Hébert, A.; Shany, E.; Richardson, J.; Geva, N.; Mawad, W.; et al. Postnatal Steroids in Preterm Infants: A Narrative Review Series—Part 2: Cardiovascular Impacts. Children 2026, 13, 395. https://doi.org/10.3390/children13030395".

Dynamic intracavitary and LVOT flow acceleration: physiology

The obstruction in steroid-associated hypertrophy is dynamic, not fixed. Its severity depends on loading conditions from beat to beat. Late in systole, the thickened walls come together and cause mid-cavity or outflow narrowing. Flow then accelerates through the narrowed segment.

Two Doppler clues support this physiology:

• The dagger-shaped, late-peaking envelope: velocity rises progressively through systole as the cavity narrows. This contrasts with the early-peaking, rounded envelope of fixed valvular stenosis.

• Mid-systolic notching: flow decelerates early in systole as the cavity closes and the gradient builds. This produces a notch in the LVOT or aortic Doppler signal.

Anything that makes the cavity smaller or more hyperdynamic worsens the gradient:

• Tachycardia shortens diastole and reduces filling. This is even more important when relaxation is already impaired.

• Hypovolemia (dehydration, excessive diuresis, bleeding, capillary leak in sepsis).

• Inotropes and chronotropes (dobutamine, epinephrine, dopamine), which increase contractility and heart rate.

• Vasodilators (including milrinone), which lower afterload and let the cavity empty further.

• Agitation and pain, through endogenous catecholamines.

Anything that makes the cavity larger or slows the heart improves it: adequate preload, a lower heart rate with a longer diastole, and higher afterload, which keeps the cavity open at end-systole.

This is the reverse of the usual reflex in neonatal hypotension. A hypotensive infant with this phenotype who receives dobutamine or epinephrine may get worse: more obstruction, lower stroke volume and more hypotension. This is the main reason to document the phenotype on TNE before the infant becomes unstable.

Reading the diastolic indices in a 26-week infant

A mitral E/A ratio below 1 is common in extremely preterm infants, even with normal diastolic function. The immature myocardium relaxes less effectively, so ventricular filling depends more on atrial contraction. A low E/A is therefore not specific in isolation at this age. In this infant, however, a reversed E/A in both ventricles, a hypertrophied, non-compliant myocardium and a small LV cavity together support a component of impaired relaxation and restrictive filling. The practical consequence is that these ventricles are highly dependent on atrial contraction and on diastolic filling time. Tachycardia and loss of sinus rhythm will be poorly tolerated. The pulmonary venous Doppler also changed. At baseline, the high D-wave reflected ductal overcirculation. On day 10, it no longer showed the high-flow pattern, consistent with ductal closure. Serial pulmonary venous and mitral Doppler can help distinguish volume loading (high D-wave, dilated LA) from impaired LV compliance (small LV, prominent atrial reversal, LA dilation without a shunt).

Management approach

The infant was stable, normotensive without support and improving from a respiratory standpoint. So the goal was not to treat an acute crisis. It was to anticipate one and plan how the infant would be managed if she became unstable (for example with sepsis, NEC, a surgical procedure or dehydration). Reassess the steroid course: the hypertrophy is expected to regress once dexamethasone is weaned. The infant was doing relatively well from a pulmonary standpoint, so the team considered shortening the course to limit further hypertrophy. The benefit of steroids in facilitating extubation and reducing BPD has to be balanced against their cardiac (and neurodevelopmental) effects. Here, the team weaned dexamethasone back to standard DART steps. Avoid tachycardia: treat pain, agitation and fever. Avoid chronotropic medications when possible. A longer diastolic filling time is protective. Avoid hypovolemia: be cautious with fluid restriction and diuretics. Hypovolemia can reveal or worsen dynamic obstruction.

If hypotension or poor systemic perfusion develops (see section on Cardiac Hypertrophy):

1. Correct tachycardia and chronotropic triggers first.

2. Give a careful volume bolus. Preload opens the LV cavity and reduces the gradient. Balance this against the risk of fluid overload and pulmonary edema in an infant with evolving BPD.

3. Avoid inotropes (dobutamine, epinephrine), chronotropes and vasodilators such as milrinone.

4. If vasoactive support is needed, choose an agent that raises systemic vascular resistance with little chronotropic or inotropic effect. This keeps the LV cavity open and reduces the gradient. Options include vasopressin or phenylephrine; norepinephrine has a mostly alpha-adrenergic effect, but some beta-1 activity should be kept in mind.

5. Get a repeat TNE to confirm the physiology and follow the gradient and cavity size.

6. In severe or refractory dynamic obstruction, beta-blockade (for example esmolol or propranolol) can be considered with pediatric cardiology, to lower heart rate and contractility and improve filling.

Glucose control: steroid-induced hyperglycemia should be monitored and managed, as hyperinsulinemia may contribute to myocardial growth.

Follow-up: serial echocardiography to document regression of hypertrophy after steroid weaning, and screening for pulmonary hypertension at 36 weeks' corrected age if the infant still needs oxygen or respiratory support.

Evolution in this case

The TNE on day 10 of dexamethasone documented ductal closure and significant biventricular hypertrophy (posterior LV wall z-score +5), with dynamic intracavitary and LVOT flow acceleration but no fixed obstruction and no SAM. Systolic function and systemic blood flow were preserved, and there were no signs of pulmonary hypertension.

The case was discussed with the primary team. The infant remained stable on NIMV without cardiovascular support, and dexamethasone was weaned back to standard DART steps to limit further myocardial hypertrophy. The team was given an individualized plan for any future hemodynamic instability: avoid tachycardia, inotropes and dehydration, give cautious volume, and use afterload-raising agents without chronotropic effect if needed. Follow-up was planned for any clinical concern, and at 36 weeks' corrected age for pulmonary hypertension screening if she still needed respiratory support. A pediatric cardiology consultation was recommended before discharge, unless a 36-week TNE showed that all findings had normalized.

Teaching points

• The ductus and the myocardium can change very quickly in the extremely preterm infant, and the same heart can show two opposite phenotypes within ten days. At the start of dexamethasone, this infant had a dilated, volume-loaded, afterload-protected LV with a large hsPDA, pulmonary overcirculation and systemic steal. On day 10, she had a closed ductus and a small, thick-walled, hyperdynamic LV with dynamic intracavitary flow acceleration.

• Postnatal dexamethasone can cause hypertrophic cardiomyopathy within days. It is usually transient and clinically silent, but it can become hemodynamically significant, especially with prolonged or higher-dose courses. Myocardial hypertrophy should be part of the risk-benefit discussion before extending a course. When an infant on dexamethasone has an echocardiogram for any reason, wall thickness should be measured and compared with z-scores.

• Ductal closure unmasks the hypertrophy. Once the ductus closes, preload falls and afterload rises. A dilated cavity can hide wall thickening, and a small cavity reveals it.

• The obstruction is dynamic. A dagger-shaped, late-peaking envelope and mid-systolic notching of the LVOT Doppler signal are useful clues, even without SAM or obvious obstruction on 2D and color. Flow acceleration can get much worse with tachycardia, hypovolemia or inotropes.

• This phenotype reverses the usual approach to neonatal hypotension. Inotropes, chronotropes and vasodilators can make things worse. Volume, a slower heart rate and agents that raise systemic vascular resistance without chronotropy are more physiologically appropriate. Documenting the phenotype on TNE before instability allows the team to plan ahead.

• Interpret diastolic indices in the context of gestational age. A reversed E/A is common in extremely preterm infants. It is more meaningful when it appears in both ventricles, together with hypertrophy and a small cavity.

• Serial TNE should follow the steroid course. The first TNE describes the starting physiology, the second shows the effects of treatment, and later studies should confirm that the hypertrophy regresses after the steroid is stopped.

References

1. Werner JC, Sicard RE, Hansen TW, Solomon E, Cowett RM, Oh W. Hypertrophic cardiomyopathy associated with dexamethasone therapy for bronchopulmonary dysplasia. J Pediatr. 1992;120(2 Pt 1):286-291.

2. Israel BA, Sherman FS, Guthrie RD. Hypertrophic cardiomyopathy associated with dexamethasone therapy for chronic lung disease in preterm infants. Am J Perinatol. 1993;10(4):307-310.

3. Skelton R, Gill AB, Parsons JM. Cardiac effects of short course dexamethasone in preterm infants. Arch Dis Child Fetal Neonatal Ed. 1998;78(2):F133-F137.

4. Doyle LW, Davis PG, Morley CJ, McPhee A, Carlin JB; DART Study Investigators. Low-dose dexamethasone facilitates extubation among chronically ventilator-dependent infants: a multicenter, international, randomized, controlled trial. Pediatrics. 2006;117(1):75-83.

5. Plessas-Azurduy, P.; Lapointe, A.; Wutthigate, P.; Spénard, S.; Villeneuve, A.; Hébert, A.; Shany, E.; Richardson, J.; Geva, N.; Mawad, W.; et al. Postnatal Steroids in Preterm Infants: A Narrative Review Series—Part 2: Cardiovascular Impacts. Children 2026, 13, 395. https://doi.org/10.3390/children13030395

6. Remy A, Vincent M, Pastor-Diez B, Picaud JC. Late postnatal steroid treatment using oral betamethasone can help to close ductus arteriosus in extremely preterm infants who cannot be weaned from ventilation. Eur J Pediatr. 2024 Nov 28;184(1):50. doi: 10.1007/s00431-024-05840-9. PMID: 39604779; PMCID: PMC11602834.

TNE before dexamethasone exposure

TNE after dexamethasone exposure