The influence of fluid resuscitation strategy on outcomes from dengue shock syndrome: a review of the management of 691 children in 7 Southeast Asian hospitals

Abstract

Introduction: The pathognomonic feature of dengue shock syndrome (DSS) is a transient capillary leak syndrome resulting in profound intravascular volume depletion. WHO management guidelines recommend particular parenteral fluid regimens during the critical leakage phase, including synthetic colloid solutions in certain circumstances. We set out to describe the actual fluid management strategies employed in different settings and to investigate relationships with clinical outcomes.

Methods: We performed a retrospective review of paediatric DSS cases managed at seven hospitals across Malaysia, Myanmar and Vietnam. We explored the effects of both initial resuscitation (crystalloid alone or mixed crystalloid/colloid in the first 2 hours) and general management: group 1 (conservative-colloid, crystalloid only), group 2 (intermediate-colloid, colloid for 1-4 hours) or group 3 (liberal-colloid, continuous colloid for more than 4 hours) categorised according to the fluid given over the first 6 hours in clinically stable patients. We incorporated an inverse probability weighting score to adjust for potential differences in baseline severity.

Results: Among all 691 patients, respiratory compromise (HR 2.08, p=0.022), requirement for nasal continuous positive airway pressure (NCPAP)/ventilation (OR 2.34, p<0.045) and days in hospital after DSS onset (risk ratio, RR 1.33, p=0.032) were significantly worse for mixed crystalloid/colloid versus crystalloid-only initial resuscitation regimens, after adjusting for baseline severity. Among the 547/691 children who stabilised within 2 hours, although a liberal-colloid general management strategy (group 3) was associated with a reduction in recurrent shock episodes (RR 0.13, p=0.043) when compared with a conservative-colloid strategy (group 1), the risks for respiratory compromise (OR 8.84, p<0.001) and requirement for NCPAP/ventilation (OR 8.16, p<0.001) were markedly increased. Additionally, the respective costs for group 3 vs group 1 were significantly higher.

Conclusions: The study highlights the potential benefits and risks of using colloid solutions in children with DSS. Formal randomised trials could help determine the most effective and safe parenteral fluid regimens for paediatric DSS. In the meantime, prolonged use of colloid solutions may be inappropriate, especially in settings without access to respiratory support.

Keywords: Arboviruses; Dengue; Treatment.

Figure 1. The vertical axis indicates the number of patients, with each line representing all fluids used for one patient. The blue colour represents colloid solutions, the green colour represents crystalloid solutions. The red colour represents blood. The degrees of colour lightness indicate how high the fluid rate per hour (mL/kg/hour) in which darker colour is for higher rate of fluid infusion. On the horizontal axis, zero is the time of onset of DSS. Blood includes whole blood and packed red cells, as well as blood products (plasma, cryoprecipitate, platelets, fresh frozen plasma and platelet-rich plasma) when used for bleeding or serious coagulation problems. Colloids include hydroxyethyl starch solutions, gelatin, albumin and plasma/platelet-rich plasma when used for volume expansion. Crystalloids include Ringer’s lactate, Hartmann’s Sterofundin, 0.9% saline and 5% dextrose in 0.9% saline. Other fluids include glucose 10%–30% with or without added electrolytes, nutritional support, etc. DSS, dengue shock syndrome.

Figure 2. Group 1: conservative-colloid; group 2: intermediate-colloid; group 3: liberal-colloid. Note that for respiratory compromise, the number at risk excludes cases with missing data and cases with respiratory compromise within the first 6 hours from shock onset (ie, within the time frame of the overall management strategy definition).