Authors (including presenting author) :
HUI WF, CHEUNG WL, HON KL, KU SW
Affiliation :
Department of Paediatrics and Adolescent Medicine, Hong Kong Children’s Hospital
Introduction :
Electrolyte disturbances are potentially serious conditions but are oftem overlooked in critical care. Acute kidney injury (AKI) are commonly encountered among critically ill children and is associated with both morbidity and mortality. However, the relationship between the two conditions and their impact on outcomes are limited.
Objectives :
We evaulated the epidemiology of acute kidney injury and the relationship between electrolyte disturbances and tubular dysfunction among critically ill children. We also determined the role of incorporating the severity of tubular dysfunction into the diagnostic criteria of AKI for assessing the degree of renal insult.
Methodology :
A prospective study recruiting children aged 1 month to 18 years old from 6/2020 to 6/2021 admitted to the paediatric intensive care unit (PICU) of the Hong Kong Children's Hospital was conducted to study the epidemiology of AKI and electrolytes disturbances. AKI was assessed according to the KDIGO criteria. The serum electrolytes profiles on sodium, potassium, calcium, phosphate and magnesium level were collected. Appropriate simultaneous urinary investigations for tubular function were performed among children with electrolyte disturbances. We then devised a staging system based on the number of types of tubular dysfunction and incorporated the system into the current AKI diagnostic criteria to assess its performance for predicting PICU outcomes.
Result & Outcome :
There were 254 episodes of admission enrolled for final analysis. The mean age was 6.4±5.5 years old and 58.3% were male. The incidence of AKI was 41.7% (stage 1: 18.5%, stage 2: 14.2% and stage 3: 9.1%). 94.9% of children developed electrolyte disturbances and 71.3% of children required additional electrolytes supplement during PICU stay. The median number of electrolytes disturbance was 3 (2) types. Hypophosphataemia (74.2%), hypocalcaemia (70.0%) and hypermagnesaemia (52.9%) were the three types of disturbances with highest incidences. A high proportion of urinary wasting of magnesium (70.6%), phosphate (67.4%) and potassium (28.6%) were detected among children with hypomagnesaemia, hypophosphataemia and hypokalaemia. 62.6% of children with ≥2 types of electrolyte disturbances showed elevated urinary beta-2-microglobulin level. Electrolyte disturbances were associated with increased duration of ventilation (p< 0.001), PICU length of stay (LOS) (p< 0.001) and mortality (p< 0.001), and the number of types of tubular dysfunction was also associated with PICU LOS (relative risk 1.6 [1.4, 1.8]). Tubular dysfunction occurred independent of AKI status with no indices showing significant difference between children with and without AKI. A tubular dysfunction staging was devised (Figure 1) showing the distribution of nil: 17.7%, stage 1: 45.3%, stage 2: 29.5% and stage 3: 7.5%. A more severe stage was associated with a significantly higher risk of prolonged PICU stay. Compared to using the AKI staging alone, incorporating the tubular dysfunction staging into the AKI staging system performed better for modeling the PICU LOS (akaike information criterion 1957.8 vs 2106.5) but less well for the prediction of PICU mortality (area under the curve of 0.893 vs 0.937).
