In this paper, we have explored possible mechanisms for the excess death rate among children randomized to receive rapid boluses of 20 to 40 ml/kg of 5% albumin or 0.9% saline fluid resuscitation compared with no-bolus controls. We found no evidence that excess 48-hour mortality associated with boluses differed by type of PS, by individual constituent components of each syndrome, or by baseline hemoglobin level. Remarkably, in every subgroup we examined, there was consistent evidence of harm by boluses. Paradoxically, the syndromes where most concern has been expressed over trial inclusion (respiratory or neurological alone and/or less severe shock criteria) not only had lower mortality overall, but also tended to have smaller differences between bolus and control, although care must be taken with interpretation of smaller subgroups. The only exception was hypoxia, present in a quarter of children at admission, which surprisingly appeared to be associated with significantly less harm from boluses. There appears to be no good rationale for this finding, which could have occurred by chance.
Even though this trial was conducted in settings with limited resources and no access to intensive care, the conduct of the trial complied to the highest standards of good clinical practice including adherence to intervention strategy and completeness of follow-up, 100% source document monitoring and the robust and blinded methodology for determining PS and TCE. An intention-to-treat analysis with no need for imputation for missing data minimized the likelihood of bias and underpinned the magnitude and importance of the unexpected findings of the trial and further analyses.
Noteworthy is that, consistent with global clinical experience, we observed a superior resolution of impaired perfusion by one hour in the bolus arms compared with the control arm. However, importantly, this did not translate into a superior outcome when compared with children with continued impaired perfusion: in both cases, boluses resulted in higher mortality. Mortality excess with boluses among children with and without hypoxia at baseline occurred to a similar extent irrespective of oxygen saturation status at one hour. Although de novo development of hypoxia at one hour was more common in the bolus arms, it was not associated with a significant increase in 48-hour mortality - suggesting that, if fluid boluses were causing pulmonary edema (and hypoxia), this was not a unifying mechanism for increased mortality from boluses. Moreover, there was little evidence that fluid overload was the mechanism for excess deaths with boluses from our analyses of neurological or respiratory TCEs. Overall, cardiovascular collapse was the main TCE and contributed most substantially to the excess mortality in the bolus arms compared to control, peaking at 2 to 11 hours post-bolus. Whilst it is possible that subtle effects of fluid overload on the lungs or brain could have been missed, our findings do not lend support to this hypothesis, particularly as the ERC review process was blind to randomization and used pre-specified TCE definitions.
A limitation of our trial was that we were unable to undertake invasive or point-of-care continuous monitoring to provide greater insight into TCEs, as in high-income intensive care settings. However, the availability of patient-centered variables from our bedside observations and laboratory data from most children at baseline has enabled further characterization of the trial participants into clinically relevant presentations (PS) in the context of where the trial was conducted. It has provided a more in-depth understanding of the spectrum of clinical groupings of children enrolled in the trial and the degree of adverse outcome fluid boluses had in these groups. Our bedside observations at pre-defined times following randomization, as well as adjudication of cause of death by the ERC, blind to randomized arm and using pre-specified criteria (TCE), provide more speculative data but remain informative because they are largely operator-independent. These methodologies are central to the internal validity of our analysis, which sought to minimize systematic bias.
Whereas initial improvement in circulatory status following bolus resuscitation is consistent with global clinical experience [8, 12, 19], the observation of excess subsequent cardiovascular collapse and excess mortality, even in the early responders, was only made possible because of exemplary adherence to randomized allocation, and in a trial protocol in which further boluses were given to only (very few) children developing severe hypotension . The possibility that fluid resuscitation lead to hemodilution [20, 21] in already anemic children, reducing oxygen delivery to the myocardium  and leading to ischemia and cardiac dysfunction, is largely ruled out by the lack of heterogeneity in the TCEs by hemoglobin level, and by the analysis showing excess harm with fluids across the whole range of hemoglobin values. These findings challenge the presumption that early and rapid reversal of shock by fluid resuscitation translates into longer-term survival benefits [8, 23, 24] in settings where intensive care facilities are not available. They do raise a possibility that rapid fluid resuscitation may cause adverse effects on vascular hemodynamics and myocardial performance, driving the requirement for inotropic and pressor support. Rapid restoration of microcirculatory perfusion may come at the expense of requiring other components of the sepsis bundle, including inotropes and ventilation. This possibility could be explored in future clinical and preclinical research examining the natural history of shock in patients managed by maintenance only (as in the control arm) and with fluid boluses.
Adverse effects of hyperchloremia, at the doses given in this trial, remain controversial [25–29]. Intriguingly, the most deleterious effects of boluses were in those patients with severe acidosis at baseline, the group with the least a priori equipoise regarding the potential benefits of fluid resuscitation - lending support to the notion of adverse effects of the resuscitation fluids on acid-base equilibrium. Alternatively, this may indicate lethal reperfusion injury  or a surge of cytokines  in cases with advanced shock. As previously suggested , shock may be an adaptive, time-dependent response sustaining children through a prolonged period prior to hospital admission - only to die within hours of reperfusion.
Whatever the explanation, these findings raise important questions about the pathophysiological mechanisms of shock and goal-directed management, questioning whether the protocol-driven requirement for inotropes and vasoactive drugs has been driven, in part, by aggressive fluid challenge . The 2008 Surviving Sepsis Campaign Guidelines, informed by a modified Delphi process, graded the pediatric recommendation (20 ml/kg boluses over 5 to 10 minutes up to 60 ml/kg) as 2C, indicating a weak recommendation with low quality of evidence. The pediatric studies on which these recommendations were based included only two retrospective, observational studies of initial resuscitation volume on the outcome of children with putative sepsis from a single tertiary referral center, involving 34 and 91 children respectively [19, 24]. The inclusion criteria for both studies were children who survived to intensive care unit admission, but were inotrope-dependent and had pulmonary arterial catheters in situ. Higher initial fluid boluses and early shock reversal in 9 and 24 children in the two respective studies were associated with improved global outcome. However, the study design and the patient population had major limitations in terms of survivorship bias and external validity to other settings. Other sources of evidence that were referenced to inform fluid resuscitation guidelines include dengue shock. We suggest these are largely irrelevant to the management of sepsis, because shock as a complication of dengue occurs 7 to 10 days after fever defervescence, secondary to gross intravascular leakage leading to vomiting, abdominal pain, increasingly tender hepatomegaly, narrow pulse pressure and hemoconcentration . The American College of Critical Care Medicine guideline, and similar guidelines, nevertheless have been adopted by many countries worldwide where point-of-care testing and intensive care facilities are available and are considered to be best practice. These are now being recommended as standards of care for resource-poor settings [33, 34]; indicating that the FEAST trial had limited generalizability because the adverse effects of fluid boluses were largely confined to children with malaria and/or anemia . The data presented in the original manuscript , subsequence correspondence  and this detailed sub-analysis definitely counter this interpretation. A recent systematic review assessing the evidence base for fluid resuscitation in the treatment of children with shock due to sepsis or severe infection found only 13 pediatric trials. Whilst the majority of all randomized evidence to date comes from the FEAST trial, they recommended that implementation of simple algorithms for children managed at hospitals with limited resources to ensure identification of children who maybe potentially be harmed by fluid boluses . The findings we report here raise questions over the rates, volumes and types of solutions recommended in pediatric resuscitation protocols, most of which remain untested in clinical trials.