Which fluids is best for sickle cell crisis?

Sickle cell disease (SCD) is comprised of a group of autosomal recessive haemoglobin disorders characterised by the production of abnormal haemoglobin, haemoglobin S (HbS). The common feature in SCD is the presence of the sickle gene either homozygous or heterozygous. Sickle cell anaemia (HbSS) occurs as a result of homozygous inheritance of the sickle gene and is the most common and severe form of the disease (Serjeant 1992). Other clinically significant diseases are compound heterozygous conditions, in which the sickle haemoglobin interacts with other haemoglobins, such as haemoglobin C (HbSC) or ß‐thalassaemia (HbSb+ and HbSb0) (Lane 2001a).

Acute painful or vaso‐occlusive crisis is the most common manifestation of SCD (Steinberg 2006). These painful episodes are periodic, often recurrent, and may occur throughout life; among women of child‐bearing age, vaso‐occlusive crisis tends to occur more frequently during pregnancy (Adam 1996). The range of duration of painful crises remains unclear; pain often lasts for between four and six days but may persist for weeks (Steinberg 1999). The pain rate (number of episodes per year) has been used as a proxy to identify people with poor clinical outcome as it correlates well with early death in people with sickle cell anaemia between the second and third decade of life (Platt 1991). However, in people less than 20 years of age, no correlation exists between pain and mortality. People who experience three or more acute painful episodes per year requiring treatment with parenteral opioid analgesics in a medical facility for four or more hours are considered to have a poor clinical outcome (Charache 1995).

The treatment of acute painful crises is at present predominantly symptomatic, comprising of: pharmacological agents (non‐opioid and opioid analgesics with or without adjuvants); non‐pharmacologic approaches (heat or ice packs, relaxation, acupuncture, self‐hypnosis); as well as other measures such as packed red cell transfusion, oxygen therapy, the treatment of associated infections and fluid replacement therapy (Ballas 2005; Ballas 2007; De Ceulaer 1982; Gaston 1986; Lane 2001a; ). Fluid replacement therapy is of paramount importance because dehydration from insensible water loss, reduced fluid intake and polyuria causes a reduction in plasma volume with an increase in blood viscosity, which promotes and sustains the sickling process (Adekile 1999). Hyposthenuria, the inability to concentrate urine, is the most common renal abnormality in people with SCD (Allon 1990; Kontessis 1992). This functional impairment in the concentrating mechanism arises due to hypoxia in the renal tubules and as a consequence of subclinical intravascular sickling in the hypertonic renal medulla (Keitel 1956). Hyposthenuria is most marked in HbSS (Statius van Eps 1967). As a result of hyposthenuria, people with SCD produce a higher than usual obligatory urine output and are more susceptible than normal individuals to dehydration (Saborio 1999). During episodes of illness, people with SCD often decrease their intake of fluids and may become dehydrated. While people with SCD in painful crisis may not be clinically dehydrated, their red blood cells are known to be dehydrated (Brugnara 1995).

Sickling of red blood cells is dependent on the intracellular concentration of deoxyhaemoglobin S which in turn is a function of the osmolality of the plasma. The aim of hydrating people in painful crisis is to slow or stop the sickling process by increasing the plasma volume, thereby decreasing blood viscosity and indirectly reducing red cellular dehydration as well as the intracellular concentration of Haemoglobin S. It has been suggested that routinely giving extra oral fluids in mild vaso‐occlusive crisis shortens the episode and relieves pain (Steinberg 1999). Supplemental fluids may be given either orally or intravenously, nasogastrically, subcutaneously or rectally. By inducing severe dilutional hyponatremia (abnormally low levels of blood sodium) in people with SCD using a vasopressin analogue and 5% dextrose in water (5DW) or half normal saline as fluid therapy as well as dietary salt restriction, it is possible to reduce the mean red cell haemoglobin S concentration (Rosa 1980). This, however, may be associated with severe adverse effects and is not a consistent finding (Charache 1981; Charache 1983). The use of 5DW or half normal saline as fluid therapy instead of normal saline has not been demonstrated to significantly affect the intracellular concentration of Haemoglobin S, or to prevent or abort painful crises (Charache 1983). In children and adolescents between one and one‐and‐a‐half times the estimated daily fluid maintenance requirements are recommended (Lane 2001b). Increased fluids may be needed if the individual is dehydrated or insensible losses are increased as in persistent fever, or both. The recommendations for adults is that fluid replacement is administered at an hourly rate of 250 ml for eight hours, and then reduced to 125 ml hourly if there are no signs or past history of congestive heart failure or renal failure (SCIC 2005). For those receiving opioids for pain, fluids should be given with care; opiates decrease cardiac output and fluid overload could precipitate heart failure. In people with acute chest syndrome, fluids are usually given at daily maintenance levels, as over‐hydration could cause regions of pneumonitis to develop (SCIC 2005). The clinical effectiveness of fluid replacement therapy for treating painful sickle crisis during pregnancy is however not known (Martí‐Carvajal 2009).

Treating vaso‐occlusive crisis is complex and requires multiple interventions. The routine administration of extra fluids, regardless of the individual's state of hydration, is currently advocated as a potentially beneficial adjunctive treatment in this condition. However, a systematic review of available evidence on the benefits and actual reported harms of fluid supplementation and the routes of supplementation is needed to clarify the uncertainties surrounding this intervention.

This is an update of previously published versions of this review (Okomo 2007; Okomo 2012).