Introduction: In a majority of Spanish hospitals, serum sodium levels (SNa) are determined by indirect electrode methodology (sodium in the liquid fraction of serum) divided by the total serum volume (mmol/l). To avoid over/underestimating SNa, a formula can be applied: SNa corrected for total proteins (TP)=SNa×93 divided by (99.1−(0.7×TP)). Hypoproteinaemia is frequent in patients receiving parenteral nutrition (PN), and is probably related to surgery, acute disease, and malnutrition. We have evaluated the impact of hypoproteinaemia on the frequency and degree of hyponatremia in PN patients.
Methods: A retrospective study of patients prescribed PN from 01/11/11 to 01/06/12. SNa was determined at baseline and during PN, and corrected for TP. All sodium levels were corrected for glycemia and presented in mmol/l. Patients with triglycerides >400 mg/dl were excluded. Indirect SNa (SNa) was compared to SNa as corrected for TP (TP-SNa). χ2, Students, and MannWhitney U test.
Results: 222 patients were prescribed PN, 57.2% of whom were men. The median age was 75 (6182). 47.3% were on the general surgery ward, 12% internal medicine, 12.1% oncology, and 8.6% hematology. 14.5% presented malnutrition (by BMI). Median duration of PN was 8 (514) days. Average baseline TP was 5.25 (S.D.: 0.76), with 93% presenting hypoproteinaemia (TP <6.5 g/dl). Baseline SNa: 138.1 (S.D.: 4.5) (P<0.001), TP-SNa: 134.6 (S.D.: 4.5) with a difference of 3.5 mmol/l (95% CI: 3.43.6) (P<0.001). 20% presented initial hyponatremia (SNa <135 mmol/) vs 52% (TP-SNa <135) (P<0.001). 28.7% developed SNa hyponatremia during PN, vs 64.2% with TP-SNa (P<0.001), within 4 (27) and 3 (114) days respectively (P=0.05).
Conclusions: Patients receiving parenteral nutrition often present hypoproteinaemia. Therefore, correction of indirect SNa for total proteins becomes essential, both to avoid underestimation of the number of patients with hyponatremia, as well as to correctly take into account its degree.