Assessment and management of low blood pressure in extremely preterm infants

INTRODUCTION — Cardiovascular assessment and management of blood pressure (BP) in extremely preterm (EPT) infants (gestational age [GA] <28 weeks) is challenging due to the presence of multiple disease processes, unpredictable adaptation to extrauterine life, and difficulty assessing organ perfusion. Although there are data that suggest EPT infants with low BP are at an increased risk for adverse outcomes, low BP in EPT infants with adequate perfusion may not be an independent risk factor for poor outcome. By contrast, a minority (approximately 10 to 15 percent) of EPT infants with low BP have evidence of poor perfusion (shock). These infants require immediate attention to restore adequate perfusion.

This topic will discuss the challenges of cardiovascular assessment in EPT infants and will review the approach to the evaluation and management of low BP in EPT infants with adequate perfusion. Evaluation and management of neonatal shock is discussed separately. (See "Neonatal shock: Etiology, clinical manifestations, and evaluation" and "Neonatal shock: Management".)

DEFINITIONS

●Extremely preterm (EPT) infant – Infants born <28 weeks gestational age (GA).

●Extremely low birth weight (ELBW) infant – Infants born with birth weight (BW) <1000 g.

●Adequate peripheral perfusion – Warm, pink skin with capillary refill <4 seconds and strong pulses to all extremities; good urine output; no laboratory evidence of end-organ dysfunction (eg, lactic acidosis).

●Poor peripheral perfusion – Signs of poor peripheral perfusion may include delayed capillary refill, mottling of the skin, cool extremities, weak pulses, oliguria, metabolic acidosis, and elevated serum lactate level. (See 'Assessment of perfusion' below.)

●Low BP – Unlike older patients, in whom low BP is typically defined as <5^th percentile for age and for whom interventions are often initiated at this threshold, there is not a clear numeric threshold for low BP in EPT infants at which intervention improves outcome (figure 1). Nevertheless, as BP decreases below the 10^th or 5^th percentile in EPT infants, ongoing frequent assessment of perfusion and BP is required to detect any subsequent signs of impaired perfusion. (See 'BP alone is an insufficient measure of perfusion' below and 'Assessment of perfusion' below.)

CHALLENGES — Although low blood pressure (BP) values based on published normative data are commonly observed in extremely preterm (EPT) infants with adequate perfusion, numerous factors make it difficult to determine if and when intervention for low BP is warranted. BP management is based on limited observational data, varies amongst clinical centers, and there remain large knowledge gaps [1-4]. This is particularly true for EPT infants with adequate peripheral perfusion. In contrast, a small minority (approximately 10 to 15 percent) of EPT infants with both low BP and poor perfusion are in shock and require immediate attention to restore adequate perfusion. (See "Neonatal shock: Management".)

The following sections present the available but limited evidence regarding these issues that make it challenging to determine the best practice for BP management in EPT infants:

●Physiologic changes to the cardiovascular system in the immediate postnatal period that impact BP. (See "Physiologic transition from intrauterine to extrauterine life".)

●Challenge of obtaining reliable and consistent measured BP values.

●BP evaluation and intervention:

•Inability to define a clinically worrisome BP value (ie, BP target value).

•Challenge of determining if there is a relationship between untreated low BP in EPT infants with good perfusion and poor outcomes, and whether there is a likelihood of subsequent development of shock.

•Challenge of determining the impact of therapies intended to increase BP on patient outcomes.

●Difficulty assessing organ perfusion with BP alone.

Physiological changes in BP — BP values in EPT infants vary considerably depending on the birth weight (BW), gestational age (GA), and postnatal age (figure 2 and figure 1) [5-7]. Even among infants with the same GA and postnatal age, there is considerable BP variability. Most EPT infants with BP values <10^th percentile for published normative values have adequate perfusion. These "low" BP values are usually transient, reflect expected physiologic changes following delivery, and are not a sign of an underlying disease process. In addition, the difference between the 5^th and 10^th BP percentile for EPT is only 1 to 2 mmHg during the first few postnatal days. BP increases spontaneously with increasing postnatal age at a rate of approximately 4 mmHg per day during the first 72 hours after birth [1,2,5,6]. These factors make it difficult to determine what is a normal BP, when is BP too low, increasing too slowly, or indicative of a pathological process, which cannot otherwise be discerned without other clinical data. The spontaneous rise in BP after birth also makes it difficult to assess the response to therapies administered to increase BP [6,8].

By contrast, low BP or BP that fails to rise normally associated with evidence of poor perfusion (ie, shock) is pathologic. In such situations, prompt evaluation and intervention targeted at the most likely cause is indicated. (See "Neonatal shock: Etiology, clinical manifestations, and evaluation", section on 'Etiologic classification'.)

Accuracy and reliability of BP measurements — In EPT infants, BP can be measured noninvasively with a cuff or invasively using pressure transducer on an indwelling umbilical arterial catheter (UAC). Radial arterial catheters are rarely used in this population due to the risk of distal ischemia.

Regardless of the method used, obtaining a reliable and consistent BP value is challenging in this population [9-14].

●Noninvasive measurement – Reliable noninvasive BP measurement requires attention to BP cuff size, location, and method of measurement [14]. Specifically, clinicians should use the appropriate-size cuff (figure 3) for the chosen site for obtaining the BP (upper arm versus thigh) and minimize movement of the extremity during the measurement. Most devices used for noninvasive BP measurements measure mean arterial BP (MAP) and estimate the systolic and diastolic BP based on mathematical algorithms unique to the monitor and device manufacturer. As a result, noninvasive BP values can vary significantly from device to device and when compared with those obtained through a UAC. While numerous studies have suggested MAP values obtained from a BP cuff may be higher than those obtained from a UAC, this is not universally the case and several studies have reported lower MAP values with a BP cuff than an invasive measurement from a UAC.

As a result, if noninvasive BP measurements are used routinely, clinicians need to know the correlation between values obtained from a UAC and noninvasive BP measurements obtained using the devices at their center.

●Invasive measurement – For invasive BP values obtained through a UAC, the pressure transducer needs to be appropriately leveled, a reasonable waveform needs to be obtained, and the potential impact of infusions through the catheter needs to be considered. (See "Intra-arterial catheterization for invasive monitoring: Indications, insertion techniques, and interpretation", section on 'Monitoring blood pressure'.)

Normative BP data for EPT infants are primarily based on measurements obtained from a UAC. These factors should be considered when comparing an individual infant's BP values against normative data and deciding whether intervention is indicated [12,13].

Limitations of normative BP ranges — Published nomograms for systolic, diastolic, and mean BP are available for EPT infants based upon normative data [6,7,15-17]. The 5^th or 10^th percentiles on these nomograms are often used to define abnormal BP. Another commonly applied definition is a MAP numerically equivalent to the infant's GA at birth (in weeks) (figure 2). However, we suggest not using these numerical threshold as the sole criteria for therapeutic intervention [4]. Characterizing a low BP value as abnormal or pathologic because if falls below the 5^th or 10^th percentile on these nomograms is problematic because BP values vary considerably during the first 72 hours after birth such that most EPT infants have at least one low systolic, diastolic, or mean BP below the 5^th or 10^th percentile [1-3,6,7,16,18]. Clinically, the difference between the 5^th and 10^th percentile may only be 1 to 2 mmHg. In addition, outcome data from a large prospective cohort reported that early postnatal low BP was not associated with developmental delay at 24 months corrected gestational age (GA) [19]. Thus, additional clinical parameters (eg, capillary refill) must be considered when assessing the adequacy of perfusion and the need for BP intervention. (See 'BP alone is an insufficient measure of perfusion' below and 'Assessment of perfusion' below.)

Therapeutic intervention based on proposed thresholds for low BP in EPT infants without clinical evidence of poor perfusion is not associated with improved infant outcomes (see 'Outcomes in treated versus untreated infants' below). As a result, there is no consensus on an accepted threshold for intervention for perceived low BP in EPT infants with adequate perfusion, and in fact, many experts in the field, including the author of this topic, do not think intervention for infants with adequate perfusion is necessary [18,20]. (See 'Adequate peripheral perfusion' below and 'Outcomes in treated versus untreated infants' below.)

BP alone is an insufficient measure of perfusion — In EPT infants, BP alone is not an accurate assessment of global perfusion since BP correlates poorly with organ perfusion, particularly cerebral perfusion [4,21-31]. Several observational studies have reported that cerebral blood or oxygenation based on measurements using near infrared spectroscopy or Doppler sonography were independent of BP changes [22,24,25,27-31]. In addition, a study of 485 infants (mean GA 28.4 weeks) reported BP measurements correlated poorly with left ventricular output [32]. (See 'Assessment of end-organ perfusion' below.)

ASSESSMENT OF PERFUSION

General assessment — For all extremely preterm (EPT) infants, we monitor blood pressure (BP) measurements either continuously through an umbilical arterial catheter (UAC) or frequently noninvasively with BP cuff (typically hourly) in conjunction with ongoing assessment of peripheral perfusion, acid-base status, and urine output. This general assessment guides further management decisions based primarily on the adequacy of peripheral perfusion and not reliance solely on BP values. (See 'BP alone is an insufficient measure of perfusion' above.)

Assessment of end-organ perfusion — In EPT infants, BP and clinical assessment of peripheral perfusion assessment do not reliably reflect end-organ perfusion (eg, cerebral blood flow). As a result, there are ongoing efforts to identify better methods for assessing end-organ perfusion. However, these have not been routinely incorporated into the management of EPT infants in most neonatal intensive care units (NICUs).

These include the following:

●Echocardiography – Echocardiography is a readily available noninvasive tool for the care of extremely preterm infants that can evaluate cardiac function, the presence of congenital heart disease, severity of pulmonary arterial hypertension, and shunting of blood through atrial, ventricular, or great vessel channels [33,34]. Serial echocardiographic studies, often performed by neonatologists, can assess changes in cardiac function over time and response to therapy, and can help differentiate causes of shock [35,36]. Previous concerns regarding inter- and intra-observer variability and unintended adverse effects (eg, induced pulmonary arterial hypertension) have diminished [36,37]. However, neither routine use of neonatologist-performed echocardiography to guide management nor therapies intended to improve echocardiography measurements of cardiac function have been shown to improve patient outcomes [35,38-40].

●Near infrared spectroscopy (NIRS) – NIRS is a direct, continuous, noninvasive measure of the oxygen balance of a specific organ. It is most commonly used to measure cerebral oxygenation and perfusion (ie, cerebral oximetry). Although NIRS appears safe with little evidence of adverse effects, data are insufficient to recommend its routine use for the care of EPT infants.

Available evidence does not demonstrate clear clinical benefits with NIRS. In a multicenter randomized trial (COSGOD III) that included 607 preterm infants, treatment guided by cerebral NIRS during the first 15 minutes after birth did not improve survival without cerebral injury compared with standard care (83 versus 79 percent) [41]. In a series of multicenter prospective studies (SafeBoosC), cerebral NIRS monitoring decreased the burden of both cerebral hyperoxia and hypoxia compared with standard care; although this study was not designed to evaluate patient outcomes, both in-hospital and two-year neurodevelopmental outcomes were similar between study patients and controls [42,43]. Similarly, in a subsequent randomized trial of 1601 EPT infants, treatment guided by cerebral NIRS initiated within six hours of birth did not reduce the rates of death, severe brain injury, or serious adverse events at 36 weeks postmenstrual age compared with usual care [44].

●Other modalities – Other modalities for investigating perfusion in EPT infants, such as non-invasive continuous measures of cardiac output through electrical cardiometry and pulsatility index show promise, but remain investigational due to limited safety and efficacy data [38,45-47].

Indicators of pending shock — Any of the following findings should raise concern that the neonate may progress to a shock state with poor end-organ perfusion. Urgent evaluation and treatment are required in these circumstances (see "Neonatal shock: Management"):

●BP fails to rise spontaneously or drops over time (figure 2 and figure 1). (See 'Physiological changes in BP' above.)

●Clinical signs of decreased peripheral perfusion such as cool extremities, acrocyanosis, and pallor. Environmental factors such as hypothermia can have an effect on perfusion in extremely preterm infants and should be promptly identified and addressed. (See "Neonatal shock: Etiology, clinical manifestations, and evaluation", section on 'Decreased peripheral perfusion'.)

●Laboratory abnormalities indicative of poor global perfusion (ie, metabolic acidosis with low serum/plasma bicarbonate and elevated blood lactate level). (See "Neonatal shock: Etiology, clinical manifestations, and evaluation", section on 'Laboratory findings'.)

MANAGEMENT APPROACH

Poor peripheral perfusion — Immediate intervention is indicated for patients with clinical signs of poor perfusion regardless of the blood pressure (BP) as these infants are in shock. (See "Neonatal shock: Etiology, clinical manifestations, and evaluation", section on 'Initial stabilization' and "Neonatal shock: Etiology, clinical manifestations, and evaluation", section on 'Clinical manifestations' and "Neonatal shock: Management", section on 'Initial stabilization'.)

The following are clinical signs of poor perfusion:

●Cool extremities.

●Pallor.

●Mottled appearance.

●Delay in capillary refill >4 seconds.

●Oliguria (urine output <1 mL/kg/hr for at least 12 hours).

●Metabolic or lactic acidosis without a known specific etiology (eg, inborn error of metabolism due to mitochondrial disorders, glycogen storage disease or disorders of gluconeogenesis). (See "Inborn errors of metabolism: Identifying the specific disorder", section on 'Lactate and pyruvate'.)

Adequate peripheral perfusion — Our suggested approach to managing extremely preterm (EPT) infants with low BP who have adequate perfusion is as follows:

●Confirm reliability of the BP measurement, as technical problems can occur and impact decision-making. The method used for BP measurements should be considered when interpreting the BP value because noninvasive cuff BP measurements (systolic, diastolic, and mean arterial pressure) can be higher than BP values obtained from an umbilical arterial catheter (UAC). (See 'Accuracy and reliability of BP measurements' above.)

●If the BP is low but perfusion is adequate, we suggest close observation rather than providing therapeutic interventions aimed at increasing BP (eg, isotonic fluid boluses, inotropes, corticosteroids). This is because the available evidence has not demonstrated a clear benefit of these interventions in EPT infants with low BP and there is some concern that they may be harmful. These data are discussed below. (See 'Outcomes in treated versus untreated infants' below.)

●Continue to monitor closely for impending signs of poor perfusion. Appropriate monitoring includes:

•Hourly or continuous BP monitoring.

•Clinical assessment and physical examination every two to three hours at a minimum to assess peripheral perfusion.

•Measuring urine output every three to four hours. (See "Neonatal acute kidney injury: Pathogenesis, etiology, clinical presentation, and diagnosis", section on 'Time of first void and urine volume'.)

•Serial blood testing to promptly identify changes in acid/base balance (blood gas) or the hemoglobin/hematocrit level.

●For most infants, no further evaluation or invention is required if their clinical course remains stable.

●However, intervention is generally warranted if there are signs of evolving neonatal shock such as the following (see 'Indicators of pending shock' above and "Neonatal shock: Etiology, clinical manifestations, and evaluation"):

•Sustained decrease in the MAP by more than 5 mmHg that is not related to technical problems or a change in how BP values are obtained.

•Failure to observe the expected rise in BP with increasing postnatal age over the first 48 to 72 hours.

•Subsequent clinical evidence of poor perfusion (delayed capillary refill, mottling of the skin, cool extremities, weak pulses, oliguria, elevated lactate level). (See 'Assessment of perfusion' above.)

OUTCOMES IN TREATED VERSUS UNTREATED INFANTS — Interventions that are used to treat low blood pressure (BP) in extremely preterm (EPT) infants include isotonic fluid boluses, inotropes (eg, dopamine, dobutamine, epinephrine), and corticosteroids (eg, hydrocortisone). For EPT infants who have adequate perfusion, it remains unclear whether these interventions have a clinically meaningful impact, and if so, whether they are beneficial or harmful [5,8,19,20,22,23,48-51]. The data addressing these questions are discussed below.

The available evidence should be interpreted with caution as it is observational in nature with notable limitations, including small sample size, inclusion of potential confounding variables such as severity of illness and the presence of underlying diseases that can lead to both shock and death (eg, cardiogenic or hypovolemic shock), and heterogeneity of study design with varying definitions of low BP, criteria for intervention, and choice of intervention.

The controversy regarding the impact of BP treatment is illustrated by the following:

●Potential harm – The following poor outcomes have been reported to be associated with interventions used for low BP in observational studies:

•Mortality [20,49,52,53].

•Intraventricular and periventricular hemorrhage [8,52-56].

•Retinopathy of prematurity (ROP) [57,58].

•Bronchopulmonary dysplasia (BPD) [54].

•Necrotizing enterocolitis (NEC) [59,60].

•Deafness [15].

•Impaired neurodevelopment [48,61].

In a prospective observational study of 367 EPT infants from the National Institute of Child Health and Human Development Neonatal Research Network, 55 percent received at least one of the following interventions during the first 24 hours after birth: fluid bolus, vasopressor, blood transfusion, corticosteroid therapy [20]. These interventions were most commonly administered to infants with ≥3 low BP values, although they were also commonly administered to infants without documented low BP (28 to 41 percent). Analyses were performed with 14 different definitions of low BP. In a univariate analysis, rates of severe ROP, grade III/IV intraventricular hemorrhage (IVH), and hospital mortality were higher in infants who received ≥1 of these interventions compared with untreated infants. However, these differences were no longer significant after controlling for study center, gestational age (GA), severity of illness, and the number of low BP values.

In a subsequent report of 331 infants (90 percent of the original cohort), the risk of death or neurodevelopmental impairment at age 18 to 22 months was higher among infants who received one of these interventions in the initial 24 hours compared with untreated infants after controlling for confounding factors (odds ratio 1.84; 95% CI 1.09-3.09) [48].

●Potential benefit – In another prospective study of 456 EPT infants, propensity score matching was used to evaluate outcomes in neonates with low BP who received intervention (fluid bolus, vasoactive infusion, or corticosteroids) during the initial three days after birth (n = 119) compared with neonates who had low BP but did not receive intervention (n = 119) [50]. Hospital mortality was similar in both groups but treated infants had a lower rate of grade III/IV IVH compared with untreated infants.

●No effect – Other smaller observational studies suggest that patients with untreated low BP (however defined) for EPT infants with adequate perfusion (capillary refill <3 seconds, strong pulse, warm extremities) appear to have similar outcomes to those with normal BP [5,23,51,62].

SUMMARY AND RECOMMENDATIONS

●Challenges – There is lack of consensus regarding the definition for low blood pressure (BP) in extremely preterm (EPT) infants (gestational age [GA] <28 weeks) and optimal management of low BP in this population is uncertain. Specific challenges include (see 'Challenges' above):

•Physiologic changes – BP values in EPT infants vary considerably depending on the birth weight (BW), GA, and postnatal age (figure 2 and figure 1 and figure 4). (See 'Physiological changes in BP' above.)

•Difficulty in obtaining reliable measurements – BP can be measured using a transducer on an indwelling umbilical artery catheter (UAC) or with a noninvasive BP cuff. For both methods, it is important to ensure that measurement is performed appropriately so that BP values are reliable and consistent. Noninvasive measurements may be higher than values obtained from a UAC. (See 'Accuracy and reliability of BP measurements' above.)

•Limitations of normative BP ranges – Most EPT infants have at least one systolic BP, diastolic BP, or mean arterial pressure (MAP) below the 5^th or 10^th percentile during the first 72 hours after birth. Thus, characterizing a BP value as pathologic because if falls below these thresholds is problematic. (See 'Limitations of normative BP ranges' above.)

•BP is an insufficient measure of perfusion – In EPT infants, BP alone is not an accurate assessment of global perfusion since BP correlates poorly with organ perfusion, particularly cerebral perfusion. (See 'BP alone is an insufficient measure of perfusion' above and 'Assessment of end-organ perfusion' above.)

●Distinguishing normal physiologic changes from shock – For most EPT infants, BP values that are lower than published normative values usually are transient, reflect normal physiologic changes following delivery, and are not a sign of an underlying disease process. (See 'Assessment of perfusion' above.)

However, infants with low BP and evidence of poor peripheral perfusion (delayed capillary refill, mottling of the skin, cool extremities, weak pulses, oliguria, metabolic acidosis, and elevated lactate level) are in shock and require urgent intervention to restore adequate perfusion. Management of neonatal shock is directed at the underlying cause, as discussed separately. (See "Neonatal shock: Management".)

●Management of low BP with adequate perfusion – Our suggested approach to managing EPT infants with low BP who have adequate perfusion is as follows (see 'Adequate peripheral perfusion' above):

•Confirm that the BP measurement is accurate. (See 'Accuracy and reliability of BP measurements' above.)

•For most EPT infants, we suggest close observation (watchful waiting) rather than providing therapeutic interventions to increase BP (eg, fluid boluses, inotropes, corticosteroids) (Grade 2C). This is because the available evidence has not demonstrated a clear benefit of these interventions in EPT infants with low BP and there is some concern that they may be harmful. (see 'Adequate peripheral perfusion' above and 'Outcomes in treated versus untreated infants' above).

Appropriate monitoring includes:

-Hourly or continuous BP monitoring

-Frequent clinical assessment of peripheral perfusion every two to three hours

-Measurement of urine output every three to four hours

-Serial laboratory testing to assess acid/base status and hemoglobin/hematocrit levels

•In most cases, no further evaluation or invention is required if the clinical course remains stable.

•However, prompt intervention is generally warranted if there are signs of evolving neonatal shock, including any of the following (see 'Indicators of pending shock' above and "Neonatal shock: Management"):

-Development of clinical signs of poor perfusion (delayed capillary refill, mottling of the skin, cool extremities, weak pulses, oliguria, elevated lactate level) (see 'Assessment of perfusion' above).

-Sustained drop in the MAP by >5 mmHg.

-Failure of BP to rise as expected with increasing postnatal age.