Poor return rate/high dropout rate requiring alteration to study methodology

Oral sodium supplementation is currently administered in cases of poor weight gain in infants particularly in patients who have undergone gastrointestinal surgical procedures. The decision to start oral supplementation is based on urinary sodium levels although the level at which to start treatment is variable as the range in normal, healthy infants is unknown. This study aims to ascertain the normal range of sodium in urine specimens collected from healthy newborn babies.

It is believed that by increasing the level of salt in the intestine, glucose can be more easily be absorbed and therefore weight gain improved. Babies with conditions where they are at risk of salt depletion (i.e. those with a stoma) are currently administered oral sodium supplementation if they are failing to gain weight and have an associated 'low' urinary sodium level. Low levels of urinary sodium are considered to represent a state of low body sodium levels, as the kidneys attempt to reabsorb most of the sodium in the urine before it is excreted.

Current practice varies widely as to the level below which treatment should be instigated. Some centres advise below 20 mmol/L (Birmingham Children's Hospital and Nottingham Children's Hospital guidelines), others below 10 mmol/L (University Children's Hospital, Zurich). In Glasgow, babies with poor weight gain are given sodium supplementation if urinary sodium levels are below 40 mmol/L. There are no documented ranges for the levels of urinary sodium in healthy, newborn babies.

By determining the reference range of urinary sodium levels in healthy, term babies who are gaining weight appropriately, the investigators hope to be able to have a better understanding about both the level below which supplementation should be considered and the target range that should be aimed for.

Study design

The study has been designed as a cross sectional study where participants details are recorded at birth, followed by a second time point at 6 weeks of age when a corresponding sample of urine will be obtained. The population being studied is healthy term (37-41 week gestation) babies with no significant ante- or post-natal history born within the study period in a single maternity unit. Participants will be required to collect a sample of urine at 6 weeks post-natal age (they will be provided with a standardized means to collect the specimen) and return it to the investigators in a pre-paid, pre-addressed envelope (UN602 compliant packaging). This design has been chosen to obtain an unbiased sample group, with minimal inconvenience or risk to participants. All confounding factors will be measured on a data-sheet accompanying the specimen (sex, weight, method of feeding and medications). The study should not be susceptible to bias as there is no perceived association between those participants who either wish to take part in the study or subsequently choose to collect and return the specimen, and their baby's urinary sodium level.

Prospective participants will be identified and approached by the neonatal/maternity staff at the pre-discharge routine baby check (see Inclusion/Exclusion criteria) from the starting date of the study for a period of 3 months (estimate based on average newborn deliveries per week and expected return yield of urinary sodium samples i.e. participant drop-out rate). There will be no coercion to participate. Mothers will be supplied with a patient information statement and informed consent obtained. The consent will be filed in the mother's medical notes and there will be no means of identification on the consent form that will link it with any specimen obtained from the mother's baby. Instead, it will be a generic form indicating intent to participate in the study. If mothers do wish to take part in the study, an initial data-set will be obtained that will be identified by a unique study participation number. This will include sex of the baby, birth weight, intended method of feeding/milk and any medication. A study pack will be issued containing a set of instructions outlining the study, a gender-specific urine collection bag, a specimen pot labeled with the participant's unique identifying number (into which the urine sample should be transferred prior to returning it), the data-set collection form and pre-paid, pre-addressed envelope (UN602 compliant packaging) in which to return the specimen and completed form.

It will be made clear to all participants that they are free to withdraw from the study for any reason at any stage and that this will not affect their future care. No attempts will be made to contact the families if samples are not returned after the 6 week time period.

There is no planned interim analysis. The study will continue until sufficient samples have been returned in order to complete meaningful statistical analysis. It is anticipated that this study should last no longer than 3 months.

Safety analysis

Participation in the study is not deemed to involve any risk for patients, their families or staff involved.

Sample size and estimation

135 participants recruited over 3 months. An independent statistician has been consulted. The sample size or reference limits has been based upon the precision of the confidence intervals around each limit. On computing a 95% reference interval based on a normal distribution (with reference limits to be at most 15% of the reference interval size) gives a sample size of 135.

A dropout rate of at least 25% is expected, and therefore, in order to achieve an adequate sample size, we intend to pause the study after 200 participants have enrolled. If less than 135 samples are returned within 6 weeks after the final participant is enrolled, then further participants will be recruited to achieve the desired sample size.

Sample analysis

Urinary electrolytes will be measured in the Biochemistry laboratory in the Royal Hospital for Sick Children. Urinary electrolytes (potassium, sodium and chloride) are measured by in-direct (diluted) ion-selective electrodes (ISEs) which utilize membranes selective for each ion species. A set volume of urine (15 ul) is diluted with diluent. A measuring electrode is then submersed in this solution. An electrical potential (voltage) is developed across the membranes between the reference and measuring electrodes in accordance with the Nernst equation. The measured voltage is compared to previously determined calibrator voltages and converted into ion concentrations (mmol/L).

Following analysis samples will be disposed in accordance with the Human Tissue Authority's Code of Practice.

Statistical analysis

The variation of urinary sodium for the cohort will be presented as a normal range.

The urinary sodium of children who have not achieved a satisfactory weight gain will be compared with those who have by a standard t test (as long as the urinary sodium values are normally distributed).

Study closure

The study will end if any one of the following criteria is met:

• The planned sample size has been achieved
• There is insufficient funding to support further recruitment, and no reasonable prospect of additional support being obtained

Ethical Consideration

The study will be carried out in accordance with the World Medical Association Declaration of Helsinki (1964) and its revisions (Tokyo [1975], Venice [1983], Hong Kong [1989], South Africa [1996] and Edinburgh[2000]). Favourable ethical opinion will be sought from an appropriate Research and Ethics Committee (REC) before patients are entered into this study. Patients will only be allowed to enter the study once either they have provided written informed consent or their next of kin have provided written informed consent. The Chief Investigator (CI) will be responsible for updating the Ethics committee of any new information related to the study.

Finance and indemnity

This study is currently not sponsored. As the study is not deemed to confer risk to participants, their families or staff, indemnity has not been sought.

Dissemination of results

It is not the intent of the researchers to inform the participants of the results of the study. Firstly, all of the specimens collected and analyzed during the study are unidentifiable there will be no way of contacting those who have taken part in the research project. Secondly, the question that the study is designed to answer has no direct impact on the care of any of the participants. Instead, the researchers aim to present their results at a specialty-specific conference (e.g. British Association of Paediatric Surgeons annual congress) and in an appropriate peer-reviewed journal, regardless of the outcome of the results.

Inclusion criteria

• Term (37-41 week gestation) infant
• Born at specified maternity unit within study period
• Uncomplicated pregnancy with no antenatal diagnosis
• Uncomplicated postnatal period: no admission to neonatal/special care unit
• Full enteral milk feeds since birth

Exclusion criteria

• < 37/40 gestation or > 41/40 gestation infants
• Antenatal diagnosis of congenital abnormality
• Postnatal admission to the neonatal unit/special care baby unit
• Period of total parenteral nutrition

The population being studied is healthy term (37-41 week gestation) babies with no significant ante- or post-natal history born within the study period in a single maternity unit.