Perioperative intravenous fluid prescribing: a multi-centre audit

BACKGROUND Excessive or inadequate intravenous fluid given in the perioperative period can affect outcomes. A number of guidelines exist but these can conflict with the entrenched practice, evidence base and prescriber knowledge. We conducted a multi-centre audit of intraoperative and postoperative intravenous fluid therapy to investigate fluid administration practice and frequency of postoperative electrolyte disturbances. METHODS A retrospective audit was done in five hospitals of adult patients undergoing elective major abdominal, gastrointestinal tract or orthopaedic surgery. The type, volume and quantity of fluid and electrolytes administered during surgery and in 3 days postoperatively was calculated, and electrolyte disturbances were studied using clinical records. RESULTS Data from four hundred thirty-one patients in five hospitals covering 1157 intravenous fluid days were collected. Balanced crystalloid solutions were almost universally used in the operating theatre and were also the most common fluid administered postoperatively, followed by hypotonic dextrose-saline solutions and 0.9 % sodium chloride. For three common uncomplicated elective operations, the volume of fluid administered intraoperatively demonstrated considerable variability. Over half of the patients received no postoperative fluid on day 1, and even more were commenced on free oral fluids immediately postoperatively or on day 1. Postoperative quantities of sodium exceeded the recommended amounts for maintenance in half of the patients who continued to receive intravenous fluids. Potassium administration in those receiving intravenous fluids was almost universally inadequate. Hypokalaemia and hyponatraemia were the common findings. CONCLUSIONS We documented the current clinical practice and confirmed that early free oral fluids and cessation of any intravenous fluids is common postoperatively in keeping with the aims of enhanced recovery after surgery programmes. Excessive sodium and water and inadequate potassium in those given intravenous fluids postoperatively is common and needs to be investigated. The variation in intraoperative fluid volume administration for three common procedures is considerable and in keeping with other international studies. Future trials of fluid therapy should include the intraoperative and postoperative phases.