Optimal fluid therapy in severe falciparum malaria has not been well defined, especially in resource poor settings where access to mechanical ventilation is limited. Recent studies suggest that liberal fluid resuscitation is harmful for severe malaria patients despite they often being hypovolemic on admission. In order to elucidate the minimum fluid therapy required to prevent complications in severe malaria, we will conduct a prospective observational study in adults with severe malaria, and also in adults with severe sepsis as a comparison group. The objective of this study is to describe the association between hemodynamic variations in conventional fluid management and the probability of developing acute kidney injury (AKI) or pulmonary edema in adults with severe malaria and severe sepsis. Hemodynamic measurements will be obtained by using transpulmonary thermodilution and arterial pulse contour analysis.

Background

Fluid therapy in severe malaria and severe sepsis.

Severe falciparum malaria causes multiple organ dysfunction including metabolic acidosis, coma, anemia, acute kidney injury (AKI), and pulmonary edema. The mortality rate of the disease is still around 15-23% despite optimal antimalarial therapy. Supportive therapy and intensive care are crucial elements in the treatment of the multiple complications of the disease. Optimal fluid therapy however, one of the fundamental elements of the supportive therapy, has not been well defined. A large randomized controlled trial of fluid therapy in shocked African children (the FEAST study) showed that fluid bolus resuscitation increased the mortality in severe malaria. A retrospective study in adult patients with severe malaria showed that fluid loading had no effect on acid-base status or renal failure. A recent study by our group showed that liberal fluid management guided by invasive monitoring was not associated with improved renal function or acid-base status, but did aggravate pulmonary edema in patients developing pulmonary capillary leakage during admission in the intensive care unit. The study found that 38% of severe malaria patients developed clinical pulmonary edema, 80% of which occurred after liberal fluid resuscitation. In this study, plasma lactate as a crude measure of tissue hypoperfusion correlated with the degree of sequestration (directly observed by OPS imaging), and not with the volume of fluid resuscitation. It was concluded that liberal fluid management is not indicated in adult severe malaria. However, this same study showed that all patients with severe malaria present with intravascular dehydration, and often have not been able to maintain proper fluid intake for a considerable time. Failure to give these patients enough fluid therapy could therefore be expected to precipitate renal failure and tissue hypoperfusion. Overall, the evidence suggests that in severe malaria, liberal fluid resuscitation is harmful despite patients often being hypovolaemic on admission. The minimum fluid therapy required to prevent complications has yet to be defined. Currently, the clinician has no guidance except for the very general adage: "keep them dry".

Evidence for fluid therapy in sepsis and septic shock is also limited. Goal-directed resuscitation during the first 6 hours is recommended by the Surviving Sepsis Campaign Guideline (SSCG), based on two randomized controlled study. Fluid challenges are one of the methods used to achieve the goals, with administration of fluid boluses and initial higher volume of intravenous fluids. Due to its inclusion in guidelines1, early goal-directed therapy (EGDT) and fluid challenges have now become common clinical practice in resource-rich countries. On the other hand, the efficacy of large fluid bolus resuscitation is questioned because of low level of evidence for physiological support and lack of clinical controlled trials comparing fluid bolus therapy versus no fluid bolus therapy. The variable availability of mechanical ventilation must be considered when making recommendations for fluid resuscitation. Since access to mechanical ventilation is often limited in developing world settings, the recommendations in the SSCG are not applicable in this context without additional evaluation. Recently, the FEAST trial in East African children with compensated shock in severe malaria and sepsis revealed that fluid bolus therapy of 20 to 40 mL/kg body weight caused a dramatic increase in case fatality. Several studies have found positive fluid balance to be associated with increased mortality in sepsis or septic shock. A retrospective study (the VASST study) included 778 patients with septic shock showed that there is significant correlation between a more positive fluid balance at both time points of 12 hours and day 4 and increased mortality. In the SOAP study, multicenter prospective cohort study enrolled 1177 patients with sepsis, cumulative fluid balance within first 72 hours after the onset was an independent predictor of mortality. Also, a prospective observational study of 164 patients with septic shock reported that there was no significant difference in 90-days mortality between patients with higher fluid volume infused (> 4.0 L) and lower volume (< 4.0 L) on day 1. They concluded that initial fluid volume administered was not associated with mortality in patients with septic shock. Optimal initial fluid volume and strategy in sepsis and septic shock is thus unclear. Especially in resource-limited countries, fluid overload is more dangerous than that in resource-rich countries, because development of pulmonary edema and ALI/ARDS is almost always fatal in settings without access to mechanical ventilation. The recommendation of fluid bolus resuscitation therapy as promoted in the SSCG should therefore be evaluated cautiously in the resource-poor setting.

Rationale

As outlined above, the optimum fluid management in severe malaria and sepsis has yet to be established, particularly in resource poor settings where access to mechanical ventilation is limited. Whilst liberal fluid resuscitation appears deleterious, it is not clear what the minimum fluid requirements are; this study aims to address this question.

Procedures

The investigators will longitudinally monitor the hemodynamic parameters along with cumulative fluid administration and fluid balance during conventional fluid strategy, as practiced by the local physicians. For monitoring, we will use transpulmonary thermodilution and arterial pulse contour analysis (PiCCO-plus, Pulsion Medical System, Germany). The PiCCO system is in routine use as part of standard clinical practice to provide hemodynamic monitoring of severely ill patients admitted to ICU. It uses a central venous and arterial line to provide continuous monitoring of hemodynamic function.

Overall, the investigators believe that by monitoring the hemodynamics using the PiCCO system will give valuable new insight into the relationship between fluid management and renal function, pulmonary edema, and acid-base status in adult patients with severe malaria. Patients with severe sepsis will be recruited as a comparison group, and the volume status will be evaluated in the same manner.

Inclusion Criteria:

A. Severe Malaria

1. Any level of asexual form of P. falciparum parasitemia on blood smear

2. Severe malaria with one or more of the following:

   i. Cerebral malaria (GCS < 11). ii. Renal impairment (Creatinine > 2mg/dL or Anuria) iii. Hypoglycaemia (Glucose < 40mg/dL) iv. Systolic blood pressure < 80mmHg with cool extremities v. Pulmonary edema vi. Venous bicarbonate < 15mmol/L vii. Venous lactate > 4mmol/L

3. Age 16-65 years

4. Written informed consent obtained from patient or attending relative.

B. Severe Sepsis

1. Negative blood smear for any Plasmodium species

2. Clinical signs of infection with two of following:

   i. Heart rate > 90/min ii. Respiratory rate > 20/min iii. Body temperature >38°C or <36°C iv. White blood cell count of > 12000/μL or < 4000/μL

3. Severe sepsis with one or more of the following due to infection:

   i. Systolic blood pressure < 90mmHg despite fluid resuscitation ii. Lactate > 4mmol/L iii. Urine output < 0.5mL/kg/hour for > 2 hours despite fluid resuscitation iv. Acute lung injury with PaO2/FiO2 < 250 in the absence of pneumonia v. Acute lung injury with PaO2/FiO2 < 200 in the presence of pneumonia vi. Creatinine > 2mg/dL vii. Bilirubin > 2mg/dL viii. Platelet count < 100000/μL

4. Age 16-65 years

5. Written informed consent obtained from patient or attending relative

Exclusion Criteria:

A. Severe Malaria

1. Patient or relative unable or unwilling to give informed consent
2. Serious pre-existing disease The following exclusion criteria described below are applied to patient for whom invasive hemodynamic monitoring will be performed.
3. Spontaneous bleeding or platelet count < 30000/μL on enrollment
4. Pregnancy.
5. Contraindication or unsuitable condition for thermodilution technique

B. Severe Sepsis

1. Patient or relative unable or unwilling to give informed consent
2. Serious pre-existing disease The following exclusion criteria described below are applied to patient for whom invasive hemodynamic monitoring will be performed.
3. Spontaneous bleeding or platelet count < 30000/μL on enrollment
4. Pregnancy.
5. Contraindication or unsuitable condition for thermodilution technique

Patients with severe malaria or severe sepsis admitted to the secondary or tertiary care center