Insights into left heart dynamics in neonatal cold septic shock—A case series

Neonatal sepsis affects approximately 22 infants per 1000 live births with mortality ranging from 3 to 19%.1 Cardiovascular dysfunction complicating sepsis, increases the risk of mortality in children but large scale neonatal data are lacking.2 Septic shock is categorised as ‘warm or cold shock’. The former is characterised by vasodilatation, lower systemic vascular resistance (SVR) and increased cardiac output. Clinically, infants might be febrile, appear ‘flushed’ and have warm extremities. The latter is characterised by peripheral vasoconstriction and elevated SVR, clinically presenting as oliguria, cold extremities and widened central-peripheral temperature gap. There is limited information on the prognostic value of echocardiography in septic children. A recent meta-analysis did not identify an association between mortality and systolic function in the paediatric population, although diastolic dysfunction noted a statistically non-significant trend.3 Cold shock amongst neonatal cohorts is understudied and it is unclear whether echocardiographic findings have diagnostic or prognostic value as this requires multi-centric prospective studies performing early and serial assessments. Therefore, we present clinical and echocardiographic information about three consecutive infants with features of cold septic shock. The submission has been approved by the Monash Health ethics committee (RES-24-0000-338Q). These consecutive infants were born during a one year period (February 2023–February 2024). There were no exclusions, and the series was compiled based on the diagnosis of septic shock. This was an extremely premature female infant born at 24 weeks gestational age (GA) and weighed 686 g at birth. Her sepsis occurred on day 10 of life, and was characterised by oxygenation failure, systemic hypotension and oliguria. Echocardiographic features included low bi-ventricular output, mitral regurgitation, low time to peak velocity/right ventricular ejection time (TPV/RVET), suggesting elevated pulmonary vascular resistance, and a closed patent ductus arteriosus. Table 1 summarises the clinical and echocardiographic information. All echocardiographic parameters normalised on reassessments after passing of the acute phase approximately 72 h later. Oxygenation failure Acute kidney injury Oxygenation failure Pulmonary hypertension Acute kidney injury Oxygenation failure Pulmonary hypertension Acute kidney injury This was an extremely premature (25 weeks GA) male infant who weighed 1000 g at birth. The presentation was at day 9 of life and was similarly characterised by oxygenation failure-pulmonary hypertension, systemic hypotension, oliguria, metabolic acidosis and coagulopathy. Echocardiogram noted low biventricular output, tricuspid and mitral regurgitation, low TPV/RVET and almost complete right to left shunt through the PDA. Table 1 summarises the clinical and echocardiographic information. The infant did not survive the sepsis, passing away within 24 h. No repeat echocardiogram could be arranged. This was a female infant born at term and presented with sepsis on day one of life with cold and clammy extremities, oxygenation failure-pulmonary hypertension, coagulopathy, oliguria and metabolic acidosis. Table 1 summarises the clinical and echocardiographic information. Echocardiogram noted low biventricular output, tricuspid and mitral regurgitation, low TPV/RVET and approximately 50% right to left shunt through the PDA. All echocardiographic parameters normalised on reassessments after passing of the acute phase approximately 72 h later. In all cases, antibiotics had been started at clinical presentation and initial inotrope was started within 2 h. The initial echocardiograms in all cases were performed within 4 h of presentation and the pertinent finding was the presence of significant mitral regurgitation, alongside poor left ventricular contractility and output (Table 1 and Figure 1 [from infant 1]). The figure contains two panels; panel A notes two blue (away going) jets out of the left ventricle, one each for mitral regurgitation and the other into the aorta (left ventricular outflow tract). Panel B depicts aortic Doppler, calculating output of 155 mL/kg/min. Mitral regurgitation may be due to elevated SVR or dilated left heart (and mitral annulus), but likely a combination. Putatively, the poorly contractile left ventricle is unable to generate enough pressure to open the aortic valve for forward aortic flow due to higher SVR, causing back flow across the mitral valve (a lower pressure circuit). The net effect of reduced forward flow is reduced renal blood flow, oliguria and deranged renal function (Table 1). The echocardiographic assessments performed in these severely unwell infants were focused rather than comprehensive to include measures such as Tissue Doppler. In particular, left ventricular diastolic function could not be assessed as trans-mitral Doppler is unreliable in the presence of mitral regurgitation. Cardiovascular dysfunction frequently complicates sepsis, manifesting as systolic and/or diastolic dysfunction, failure of peripheral vasoregulation and oxygenation failure (oftentimes pulmonary hypertension). Circulatory impairment appears more severe amongst premature infants (possibly due to myocardial/vascular maturational peculiarities and relative adrenal insufficiency) and there is a higher incidence of pulmonary hypertension.4, 5 The cardiovascular response may be hyper-dynamic (warm shock: low SVR/systemic vasodilatation/high output/warm extremities) or hypo-dynamic (cold shock: higher SVR/vasoconstriction/low output/cold-pale extremities). Based on clinical symptoms and echocardiographic findings, the infants in our series likely represent cold hypo-dynamic shock. Paediatric patients with septic shock can present as either warm (hyper-dynamic) or cold (hypo-dynamic) shock, but the majority have initial high SVR and low cardiac output state, suggesting that cold shock may be the predominant presentation.6 Limited available literature describes the haemodynamic pattern of sepsis in neonates. A recent systematic review of 12 studies analysed the hemodynamic function using functional echocardiography.7 All the included studies describe hemodynamic changes of ‘warm shock’, that is, lower SVR and increased biventricular outputs. From the available data, it is unclear whether ‘cold shock’ is less common, or infants rapidly transition from cold to warm shock, prior to assessment with functional echocardiography by the point-of-care clinician or cardiology service, or receive early management based on clinical assessment alone. For instance, just 4 out of 12 studies reported the timing of echocardiography in the aforementioned review; the scans were performed at median time of 15.6 h from the time of suspicion of sepsis. Nonetheless, available data shows that infants with low SVR and adequate ventricular output are more likely to survive, while the non-survivors had lower initial cardiac output and high SVR.8 There is dearth of prospective echocardiography studies documenting the prevalence of cardiac dysfunction and its effect on outcomes in neonatal sepsis. Early and serial assessments may shed light on the transition between various phases of shock. Point of care echocardiography allows the clinician to closely study evolution of hemodynamics, and putatively tailor therapy accordingly. Arvind Sehgal: Conceptualization; investigation; writing – original draft; methodology; validation; visualization; writing – review and editing; formal analysis; data curation; supervision. Calum T. Roberts: Conceptualization; investigation; validation; visualization; writing – review and editing; formal analysis; project administration; data curation. The authors declare no competing financial interests. Arvind Sehgal Calum T. Roberts