Neonatal Septic Shock and Hemodynamic Physiology

Neonatal septic shock is a critical expression of systemic infection in which the circulation fails to meet tissue metabolic demands. In the neonatal period, sepsis-related hemodynamic instability is a major driver of mortality and organ dysfunction. Clinically, these infants typically present with a combination of cardiovascular, respiratory, and metabolic abnormalities. The integration of physiology, bedside clinical assessment, and targeted neonatal echocardiography (TnECHO) offers a powerful framework to understand and manage these complex states in an individualized way. The pathophysiology begins with a systemic inflammatory cascade triggered by infection. This involves the release of cytokines, endogenous catecholamines, nitric oxide, and other vasoactive mediators. Endothelial dysfunction and capillary leak alter vascular tone and permeability, and sepsis also has direct and indirect effects on myocardial performance and mitochondrial function. 

"Cold" vs "Warm" shock:

The balance between vasoconstrictive and vasodilatory influences, and between myocardial reserve and injury, determines how the infant presents at the bedside. Rather than a single uniform phenotype, neonatal septic shock spans a spectrum from predominantly vasoconstrictive “cold” shock to vasodilatory “warm” shock, with many infants demonstrating mixed features that evolve over time. Traditionally, cold shock refers to a state of high systemic vascular resistance and relatively low cardiac output. Clinically, these infants often have cool extremities, prolonged capillary refill, weak pulses and a narrow pulse pressure; arterial pressure may be normal or even elevated initially, despite global hypoperfusion. In contrast, warm shock is characterized by low systemic vascular resistance with normal or high cardiac output early on. These babies tend to have warm, ruddy extremities, brisk or even “flash” capillary refill, bounding pulses, a widened pulse pressure and hypotension as the syndrome progresses. In neonates, warm shock appears to be the predominant phenotype, especially in late-onset Gram-negative sepsis, although elements of both patterns can coexist in the same patient as the disease evolves.

An adequate hemodynamic assessment in this context must integrate clinical examination, biochemical markers and, where available, echocardiography. At the bedside, heart rate trends, serial blood pressures, capillary refill, peripheral temperature, urine output and lactate provide important information. In many units, noninvasive blood pressure is obtained at relatively long intervals; in a deteriorating infant with sepsis, more frequent measurements can reveal patterns such as a progressive fall in diastolic pressure, heralding evolving vasodilation before the systolic pressure declines. Tachycardia in this setting may be an early compensatory response and should not be dismissed. Elevated lactate is an indicator of a failure of oxygen delivery relative to demand and reflects the point at which the infant can no longer maintain sufficient oxygen extraction to preserve aerobic metabolism. Near-infrared spectroscopy, when available, can add continuous information about regional tissue oxygenation and perfusion.

Management

The overarching goal in managing neonatal septic shock is to restore and maintain adequate oxygen delivery to tissues so they can maintain aerobic metabolism. Blood pressure is one piece of this puzzle, but it is not synonymous with perfusion. Oxygen delivery depends on cardiac output, hemoglobin concentration and arterial oxygen saturation, and it must be matched to tissue-level oxygen consumption. Sepsis disturbs both sides of this balance, increasing demand in some tissues and reducing delivery through circulatory failure, microvascular dysfunction and impaired oxygen extraction.

Targeted neonatal echocardiography plays a role in clarifying the hemodynamic phenotype at the bedside. A comprehensive TnECHO study does not rely on a single view but acquires multiple windows and uses qualitative impressions alongside quantitative measures. The echocardiographer assesses left and right ventricular systolic and diastolic performance, estimates preload, considers afterload and systemic vascular resistance, evaluates pulmonary artery pressures, PVR/SVR relationship, ductal shunting contribution, examines the ductus arteriosus and foramen ovale for shunt direction and magnitude, and, where possible, derives stroke volume and cardiac output estimates. Parameters such as fractional area change, TAPSE, annular tissue Doppler velocities, strain or strain rate, LV and RV volumes, aortic or pulmonary VTI, and IVC size and collapsibility can all contribute to a nuanced assessment. Serial echocardiograms are particularly powerful because they reveal the trajectory of disease and the impact of therapy over time.

Warm (vasodilatory) shock is characterized hemodynamically by low systemic vascular resistance and reduced venous return. Venodilation decreases the effective stressed volume, limiting right ventricular preload and, by extension, left ventricular filling. On echocardiography, the ventricles often appear small and hyperdynamic, with nearly “kissing” endocardial surfaces in systole, reflecting vigorous contraction in the setting of inadequate filling. Stroke volume is relatively low, even if heart rate is high and the heart looks “strong.” In this situation, vasopressors are the principal pharmacologic therapy to restore vascular tone, raise venous return and support arterial pressure. Norepinephrine or vasopressin are commonly chosen agents in older populations; in neonates, practice is evolving but the physiologic rationale is similar. Dopamine has traditionally been used and may also provide some degree of inotropy as well as increased PVR/SVR ratio in those with a significant left to right PDA that may be contributive to systemic steal. Fluid boluses may be used cautiously when there is evidence of preload depletion and potential fluid responsiveness, but they should not be administered reflexively or in large cumulative volumes. If myocardial depression develops on top of vasodilation, inotropes are introduced to augment contractility.

Cold (vasoconstrictive) shock has a different physiology. High systemic vascular resistance increases afterload, making it harder for the left ventricle to eject. Over time, the ventricle may fatigue, leading to depressed systolic performance and low stroke volume. On echocardiography, the ventricles may appear poorly contractile with reduced fractional shortening or ejection fraction, and outflow tract VTI is low. These patients often have elevated blood pressure initially, with cool extremities and diminished pulses. Here, the primary therapeutic strategy is to support myocardial performance and reduce the excessive afterload. Inotropes such as epinephrine or dobutamine are preferred. Aggressive vasoconstrictors can further compromise perfusion and should be used with caution. Volume expansion is again guided by evidence of low preload rather than applied systematically.

Fluid supplementation?

The question of fluid responsiveness is central to safe resuscitation. The Frank–Starling relationship describes how stroke volume increases with preload up to an inflection point beyond which further volume does not improve output and may be harmful. A neonate on the steep portion of the curve will show a meaningful increase in stroke volume after a modest fluid bolus; a neonate already on the flat portion will not. In the latter case, additional fluid primarily contributes to interstitial and pulmonary edema, particularly in preterm infants with immature lungs and fragile capillaries. Echocardiographic tools such as IVC collapsibility or distensibility, changes in LV outflow tract VTI before and after a small fluid challenge, and visual assessment of chamber filling can help discriminate between fluid-responsive and fluid-unresponsive states. Some of these tools may be obscured by the positive pressure ventilation and/or the adapting right ventricle which is still transitioning (i.e. systemic venous return assessment by TNE). The goal is to titrate volume based on physiologic response rather than fixed protocols.

Vasoactive medications target both the arterial and venous sides of the circulation. Vasopressors increase arterial tone and diastolic pressure, but they also constrict capacitance vessels, shifting blood from the unstressed to the stressed volume, thus improving venous return and preload. Inotropes primarily enhance myocardial contractile force and may have variable effects on systemic vascular resistance. Milrinone, for example, supports contractility while reducing afterload and improving lusitropy, though its use in acute sepsis is nuanced and must account for blood pressure, coronary perfusion and renal excretion. Sepsis itself alters receptor sensitivity, intracellular signaling, and the balance between oxygen utilization and mitochondrial function, so responses to these agents can be variable.

Macro- and Microcirculation

A critical concept that spans all of these considerations is the coherence between macro- and microcirculation. It is possible to normalize systemic blood pressure and even cardiac output while microvascular flow and tissue oxygenation remain impaired. Elevated lactate is a reminder that, despite seemingly adequate global parameters, tissues may still experience oxygen debt. There may also be a component of mitochondrial dysfunction due to the infection and inflammatory storm. Before lactate rises, the organism attempts to maintain aerobic metabolism by increasing oxygen extraction at the tissue level; a high lactate indicates that this adaptive mechanism has been exhausted. In that sense, a high lactate in sepsis is both a marker of disease severity and a testimony to how hard the infant’s physiology has attempted to compensate.

Pulmonary Hypertension in sepsis

The relationship between sepsis and pulmonary hypertension in neonates deserves particular attention. Historically, early-onset sepsis has been associated with acute pulmonary hypertension and hypoxemic respiratory failure. This is secondary to acute increase in pulmonary vascular resistance or pulmonary capillary leak leading to pulmonary edema and ventilation-perf However, emerging data suggest that in late-onset sepsis, pulmonary hypertension is often not the dominant driver of hypoxemia. Instead, many infants show primary myocardial dysfunction and complex interactions between systemic vascular resistance, pulmonary vascular resistance, lung disease and shunt physiology. For example, a bidirectional ductal shunt in a septic preterm infant does not necessarily signify “primary PPHN”; it may reflect simultaneously elevated pulmonary vascular resistance due to lung pathology and low systemic vascular resistance from vasodilatory shock. In such cases, inhaled nitric oxide may not produce the expected clinical improvement because the primary limitation is not fixed pulmonary vasoconstriction. Optimizing systemic perfusion with vasopressors, treating underlying lung disease, ensuring adequate but not excessive ventilation and improving cardiac output may have a larger impact.

Putting everything together

When TnECHO is integrated thoughtfully into this context, distinct hemodynamic phenotypes emerge. A vasodilatory, warm shock pattern is represented by a small, vigorously contracting left ventricle with low filling, high heart rate and low stroke volume; the right ventricle is often hyperdynamic as well. In vasoconstrictive, cold shock, both ventricles may appear strained, with reduced systolic performance and evidence of high afterload. Mixed phenotypes show elements of both patterns and may be accompanied by variable ductal flow patterns and subtle pulmonary vascular changes. Serial echocardiograms document how these patterns evolve with time and therapy and help refine or redirect treatment, for instance, de-escalating fluids in a baby who has become fluid unresponsive or switching from predominant vasopressor therapy to more inotropic support when myocardial fatigue appears.

In summary, neonatal septic shock is not a single entity but a heterogeneous syndrome that requires individualized, physiology-based management. Clinical examination, serial vital signs, biochemical markers and TnECHO should be viewed as complementary rather than competing tools. Echo is not used in isolation or as a purely technical exercise; it is a way to visualize and quantify the circulation to support thoughtful decision-making. Rather than treating numbers in isolation, the clinician’s task is to understand how all the components of oxygen delivery and utilization fit together for a particular infant at a particular point in time.

Key practical implications follow from this approach. It is important to recognize early whether an infant is drifting toward warm or cold shock and to acknowledge that this may change as disease progresses. Volume resuscitation should be guided by evidence of fluid responsiveness, not applied in large, fixed amounts. Vasoactive agents should be selected and titrated based on the predominant physiologic disturbance, whether low SVR, high SVR, or myocardial dysfunction. Elevated lactate should trigger re-evaluation of both macro- and microcirculation. Finally, the presence or absence of pulmonary hypertension must be interpreted in the context of the overall hemodynamic picture, including ventilation, lung disease, ductal physiology and the effects of sepsis on both systemic and pulmonary circulations.

This physiologic, echo-informed framework transforms septic shock from a uniform label into a series of distinct, interpretable states. It allows clinicians to move beyond protocolized therapy toward individualized, goal-directed care, with the central aim of restoring oxygen delivery, preserving organ function and improving outcomes for critically ill neonates.

High LVO and RVO in the context of a "warm septic shock"

Teixeira-Neto FJ, Valverde A. Clinical Application of the Fluid Challenge Approach in Goal-Directed Fluid Therapy: What Can We Learn From Human Studies? Front Vet Sci. 2021 Aug 3;8:701377. doi: 10.3389/fvets.2021.701377. PMID: 34414228; PMCID: PMC8368984.