Retrospective audit of adult patients who underwent major surgery in our institution, and biochemical outcomes including hyponatremia, in relation to the maintenance fluid tonicity administered peri-operatively.

Hyponatraemia is a common electrolyte abnormality which is hospital-acquired in two-thirds of cases, and can lead to obtundation, seizures, encephalopathy and death. In children, the use of hypotonic maintenance fluids had been routine until recent evidence showing that they were associated with an increased incidence of hyponatraemia, with consequent increases in morbidity and mortality. Comparative data in adults is limited.

In our institution, the National University Hospital, Singapore, a hypotonic solution of 0.33% saline pre-mixed with 10mM KCl per 500ml has been in common use for years for surgical patients. This is largely based on calculations of daily requirements (1-1.5mM/kg of sodium and 1mM/kg of potassium) and reduction of risk associated with measuring out KCl and injecting it into drip bags. However, this solution is hypotonic (134mM) compared to plasma, and its administration could theoretically increase the risk of subsequent hyponatraemia.

In this single-centre retrospective observational study, we aim to compare the biochemical outcomes of hospitalised surgical adult patients who have been given hypotonic (0.33% saline) compared to isotonic or near-isotonic solutions (0.9% saline or Hartmann's solution) between March and September 2013. We hypothesize that hypotonic solutions are more likely to cause a post-operative hyponatraemia and subsequent complications compared to their isotonic counterparts. The complications of hyponatraemia are defined in this study as new ICU admissions and increased mortality.

Inappropriate fluid prescription is a contributor to iatrogenic morbidity and mortality. Our study aims to evaluate the safety of a hypotonic solution in standard use. The results will help guide fluid choice, especially in the absence of current evidence-based national guidelines. As it is a retrospective study, there are no risks involved.

This is a single-centre retrospective observational study based on electronic medical records collected between the months of March 2012 and September 2013. Patients will be divided into two groups - one which received hypotonic 0.33% saline, and another which received isotonic or near-isotonic solutions (0.9% saline or Hartmann's solution). Serum sodium, chloride, bicarbonate and creatinine measured pre-operatively will be compared to post-operative values obtained 24-48 hours later.

1. All medication (including fluids) administered during an inpatient stay at NUH are logged and stored electronically via the inpatient medication record system. Following a patient's discharge, this information is stored on the hospital's servers and accessible by pharmacy. By searching for a certain medication, e.g. "normal saline", a list of all patients given that medication during a certain time period can be generated, along with associated data such as when the medication was commenced and stopped.
2. From this list of patients, their IDs can be used to extract surgical data from electronic records which are automatically created every time an operative report (i.e. every time a surgical procedure is completed) is issued. Variables obtained from this database include the surgical procedure name, code, performing department, etc.
3. For biochemical data, the hospital's electronic medical records can be searched by patient IDs. Details such as documented past medical history and laboratory indices are available. By comparing dates of surgery to dates of renal panels taken for a specific patient, pre-operative and post-operative variables can be identified and extracted.

Inclusion Criteria:

• exclusive administration of either fluid types.

Exclusion Criteria:

• Patients who received saline boluses

Adult surgical patients

Request to be made to corresponding author please