We aimed in our present study to evaluate whether fluid management within the first 12 hours from the onset of septic shock in children affected the clinical outcomes. Our results indicated that the total amount of fluid given within the first 12 hours was similar between the children who survived and those that did not with no difference in the net fluid balance between these two groups. What made the difference between the survivors and nonsurvivors was the distribution of the fluid administered within the first 12 hours. Survivors tended to be given about 40% of the total administered fluids over 12 hours in the first 3 hour period with a gradual decrease in volume thereafter. However, the non-survivors showed a much more even distribution of administered fluid over the first 12 hours after septic shock. Of particular note, the higher proportion fluid group for the first 3 hour period after septic shock onset had a significantly low 28-day mortality whereas the patients in the higher proportion group for the last 3 hour period (i.e. 9–12 hours post septic shock) showed the opposite result and had a significantly high 28-day mortality.
The American College of Critical Care Medicine (ACCM) treatment guidelines for the first hour after the diagnosis of severe sepsis first and foremost recommend an aggressive fluid resuscitation of up to 60 mL/kg, followed by the titration of vasoactive medications based on the shock phenotype and, if necessary, additional fluids [4, 5, 11]. Even in the most recent SSC guidelines [6], the fluid therapy regimens have not changed. In practicality however, an aggressive resuscitation protocol of up to 60 mL/kg within 60 min is not easy to administer and it is for this reason that there is insufficient time to assess the effects or adverse effects of fluid resuscitation in pediatric cases. Moreover, increasing consideration is now being given to the mounting evidence that a significant volume overload beyond the initial period of resuscitation may be harmful [7, 8, 12]. Fluid resuscitation is one of the essential components of septic shock treatment and it is obvious that early fluid resuscitation in the initial stages of septic shock is key to a successful outcome, as indicated in many prior reports [2–4, 13–15]. In our current study of a pediatric septic shock, the survival rate was higher in the children who received more fluid in the first 3 hours. This is a very important finding in terms of the future management of these cases to improve perfusion i.e. the microcirculation in the initial period of septic shock. However it is difficult to define the appropriate fluid volumes and the optimal hemodynamic targets in children. The important determinant of the fluid resuscitation approach would be the fluid responsiveness but this is very difficult to estimate and there is no consensus on the assessment tools for this purpose in pediatric patients [16–19].
In our current cohort, a smaller volume of resuscitation fluid was administered than is recommended the ACCM guidelines, i.e. an approximately 20 ml/kg volume within the first 3 hours. Notably however, the mortality rate among our study subjects was no higher than that reported in other studies. Although our institute follows the guidelines recommended by the ACCM or SSC, the fluid resuscitation approaches in septic shock are at the discretion of the treating physicians. These clinicians consider a number of factors that could affect the patient’s fluid responsiveness, and the possible adverse effects of fluid resuscitation, including the impacts on cardiac function, respiratory function or renal function. Hence, the children in our current analysis did not receive the aggressive fluid resuscitation regimen recommended by the existing guidelines. Importantly, this did not seem to increase the mortality rate. This is another important observation from our present analysis and raises questions about whether the currently recommended aggressive fluid resuscitation approaches are essential within the first hour of septic shock onset. The goal of fluid resuscitation is to increase the mean circulating pressure and stroke volume, thereby improving the tissue perfusion pressure. However, the ability of crystalloids to expand the intravascular volume is poor and several prior studies have reported that less than 5% of a crystalloid bolus remains in the intravascular space at one hour after the end of the infusion [20, 21]. Macro-circulatory goals such as blood pressure or central venous pressure are currently recognized as poor indicators of microcirculation, especially in sepsis and septic shock [22]. Moreover, the microvasculature plays an independent role in tissue perfusion and oxygenation that may not be influenced by macrovascular alteration [23]. In relation to fluid bolus therapy, some animal studies using intra vital microscopy and video imaging of the microcirculation have shown both improvement as well as persistence in microcirculatory dysfunction with bolus therapy [24]. Thus, fluid therapy targeted at the macrocirculatory indicators can cause a fluid overload. The association of fluid overload and harm would be consistent in a broad spectrum of critically ill children. Most guidelines suggest an amount of fluid based on body weight and there is usually no accurate recommendation for fluid management after 1 hours. However, our present findings strongly indicate that basing the distribution of intake fluids over time on the patient's condition rather than body weight will have a far more beneficial impact on mortality outcomes in children with septic shock.
The recent Fluid Expansion as Supportive Therapy (FEAST) study, a Randomized Controlled Trial of fluid bolus therapy in over 3,000 acutely ill African children with sepsis and impaired perfusion, has been pivotal in generating interest in the potential harm from fluid resuscitation. The results of this trial indicated that both a higher early mortality (< 48 h) and higher late mortality (4 weeks) among children who received fluid boluses in response to impaired perfusion compared to those who did not receive fluid bolus [9]. Interestingly, in the post hoc analysis of the FEAST study, fluid boluses were found to have provided short-term benefit in terms of resolution of shock state, but the children who received these fluid boluses had increased mortality due to progression of cardiovascular dysfunction after this initial improvement [25, 26].
Abulebda et al. have described how the volume balance may influence the clinical course in pediatric patients with septic shock during the period after ICU admission [27]. These authors stratified their analysis based on the risk of mortality using their risk stratification tool and reported that an increased fluid intake and positive fluid balance after ICU admission are associated with poorer outcomes in pediatric septic patients with a low initial mortality risk, but not in patients with a moderate or high mortality risk. Their results deviate somewhat from what others have described in adults with septic shock [7, 8, 28–30] and show confounding outcomes according to the mortality risk. The reason for this limitation, as these authors pointed out, is the unavailability of data on the fluid balance of the subjects prior to their ICU admission. It is difficult to predict what the net effect on their results would have been if pre-ICU fluid balance data were available. On the other hand, a strength of our present study may be that the total intake fluid was estimated from the onset of shock regardless of the patient’s location, i.e. whether they were in the emergency room or general ward. The effects of early fluid therapy have therefore been more clearly demonstrated in our current investigation.
The existing reviews in the literature have revealed that a positive fluid balance recognized at different times during the first 24 hours in the ICU, leading to a cumulative positive balance at discharge, is predictive of a higher mortality [10, 12, 31–35]. Of note in particular, Merik et al. have demonstrated that a low volume resuscitation below that recommended by the SSC guidelines on the first ICU day in septic shock cases was associated with the reduction in mortality [33]. It is noteworthy also that our present results, in a comparable manner to previous studies, have revealed that increases in the intake volumes from 3 hours after the onset of septic shock increases mortality. Importantly, we show from our current data that a high intake volume from 9 to 12 hours post-onset has a significant association with high mortality in children with septic shock. These findings indicate that an aggressive fluid resuscitation approach from 3 hours after septic shock onset may be harmful.
Our study had several limitations of note. First, it was conducted in a single center and our results may therefore be lacking in general applicability. Second, our observational design incorporated a short period of only 12 hours, and we enrolled a smaller number of cases than other reports on fluid therapy. This limited our ability to determine a common relationship between fluid resuscitation and patient outcomes. Lastly, other factors such as antibiotic therapy, source control, and some other unmeasured clinical parameters may have contributed to our findings. For this reason, the prognosis of pediatric patients with septic shock cannot be explained simply by fluid therapy alone. It would thus appear difficult at present to standardize a single treatment protocol for these critically ill children or to draw clear conclusions from observational studies.