Myocardial and haemodynamic responses to two fluid regimens in African children with severe malnutrition and hypovolaemic shock (AFRIM study)

Abstract

Background: Fluid therapy in severely malnourished children is hypothesized to be deleterious owing to compromised cardiac function. We evaluated World Health Organization (WHO) fluid resuscitation guidelines for hypovolaemic shock using myocardial and haemodynamic function and safety endpoints.

Methods: A prospective observational study of two sequential fluid management strategies was conducted at two East African hospitals. Eligible participants were severely malnourished children, aged 6-60 months, with hypovolaemic shock secondary to gastroenteritis. Group 1 received up to two boluses of 15 ml/kg/h of Ringer's lactate (RL) prior to rehydration as per WHO guidelines. Group 2 received rehydration only (10 ml/kg/h of RL) up to a maximum of 5 h. Comprehensive clinical, haemodynamic and echocardiographic data were collected from admission to day 28.

Results: Twenty children were enrolled (11 in group 1 and 9 in group 2), including 15 children (75%) with kwashiorkor, 8 (40%) with elevated brain natriuretic peptide >300 pg/ml, and 9 (45%) with markedly elevated median systemic vascular resistance index (SVRI) >1600 dscm-⁵/m² indicative of severe hypovolaemia. Echocardiographic evidence of fluid-responsiveness (FR) was heterogeneous in group 1, with both increased and decreased stroke volume and myocardial fractional shortening. In group 2, these variables were more homogenous and typical of FR. Median SVRI marginally decreased post fluid administration (both groups) but remained high at 24 h. Mortality at 48 h and to day 28, respectively, was 36% (4 deaths) and 81.8% (9 deaths) in group 1 and 44% (4 deaths) and 55.6% (5 deaths) in group 2. We observed no pulmonary oedema or congestive cardiac failure on or during admission; most deaths were unrelated to fluid interventions or echocardiographic findings of response to fluids.

Conclusion: Baseline and cardiac response to fluid resuscitation do not indicate an effect of compromised cardiac function on response to fluid loading or that fluid overload is common in severely malnourished children with hypovolaemic shock. Endocrine response to shock and persistently high SVRI post fluid-therapy resuscitation may indicate a need for further research investigating enhanced fluid volumes to adequately correct volume deficit. The adverse outcomes are concerning, but appear to be unrelated to immediate fluid management.

Keywords: African; Children; Echocardiography; Fluid resuscitation; Gastroenteritis; Hypovolaemic shock; Mortality; Myocardial function; Ringers lactate; Severe malnutrition.

Fig. 1

AFRIM study flow

Fig. 2

Stroke volume index (SVI) (ml/min/m²) by group. The red lines indicate published upper and lower limits of stroke volume index in children below 5 years [22]

Fig. 3

a Stroke volume index (SVI) in group1 plotted against end-diastolic volume index (EDVI) and (b) SVI in group 2 plotted against EDVI; pre-fluid, post-fluid Arrows indicate the direction of change in the SVI relative to the EDVI after fluid administration. Solid circles represent survivors. Crosses represent deaths

Fig. 4

Global left-ventricle myocardial strain/deformation in (a) the radial and (b) the circumferential and (c) the longitudinal axes, by group. The red lines indicate published mean references of left ventricular cardiac strain in children from pooled meta-analysis [23]

Fig. 5

Median age-adjusted heart rate (a), systolic blood pressure (b) and diastolic blood pressure (c) by time and survival status