Prevalence and outcome of selected electrolyte abnormalities in children of 2-59 months admitted with severe pneumonia in Acute Care Unit, Mulago Hospital, Uganda
Abstract
BACKGROUND:
Pneumonia is the most common cause of morbidity and mortality in children less than five years in Uganda. Electrolyte abnormalities in children with pneumonia have been associated with increased risk of morbidity and mortality. In Uganda there is limited data on the prevalence of electrolyte abnormalities in children with severe pneumonia, which is required to guide fluid therapy in these children. This study was done to determine; the prevalence, clinical profile and outcome of sodium and bicarbonate abnormalities in children admitted with severe pneumonia in
acute care unit, Mulago hospital.
METHODS: A descriptive prospective study was conducted in Acute Care Unit, Mulago
Hospital between June and October 2014. A total of 332 children ages 2-59months with severe
pneumonia as per the WHO classification of pneumonia were enrolled. Serum sodium and
bicarbonate levels were measured at admission, and repeated after 72 hours if baseline results
where abnormal. Clinical review was done upto 48 hours, and outcome was observed until
discharge, or when the outcome of interest occurred such as death, duration of hospital stay.
Status of sodium and electrolyte abnormalities at 72hours was noted. Data was entered by the
principal investigator using Epidata version 3.1. Data was analysed using STATA version 12.0
software.
RESULTS: A total of 340 children were screened; 322 were enrolled and included in the
analysis. Hyponatraemia (serum sodium level <135mmol/l) contributed 21.2% and acidosis
(serum bicarbonate level <22mmol/l) 78.3% of the sodium and bicarbonate abnormalities among
children with severe pneumonia. Children with need for oxygen after 24 hours of admission[OR 0.48, 95% CI (0.26 - 0.91), p=0.03] were more likely to have acidosis. The median duration
of hospital stay for all participants was four days (IQR=3days). Mean duration of hospital stay for children with hyponatraemia was longer than those with normal serum sodium levels (mean
5.6days: SD ±4.83, p-value=0.02). There was no statistical difference in duration of hospital stay
between those with acidosis and those with normal bicarbonate levels (t-statistic= -0.36, df=318,
p-value=0.72). There was a large proportion of children who still had hyponatraemia (30.2%)
and acidosis (93.2%) after 72hours of admission. However, acidosis was associated with higher
mortality (6/7 =85.7%).
CONCLUSION: There is a high prevalence of hyponatraemia (21.1%) and acidosis (78.3%)
among children with severe pneumonia. Acidosis is associated with increased risk of mortality,
need for oxygen after 24hours and tachycardia. Hyponatraemia was associated with longer
hospital stay. A significant proportion of these have persistent hyponatraemia and acidosis even
after 72hours on treatment for pneumonia.
RECOMMENDATION: There is need to do routine electrolyte measurements in infants with
severe pneumonia, particularly those with tachycardia, need for oxygen 24hours after admission,
being managed for more than one diagnoses, and those with complications of pneumonia. This
will help with quick identification and management of children with acidosis, reducing duration
of hospital stay and mortality.
A guideline can be made from these results to help clinicians prioritise children for whom
electrolyte measurement should be done.