Table 3 Adjusted change in serum sodium, potassium, and chloride levels following novel furosemide or chlorothiazide administration.

From: Association of furosemide versus chlorothiazide exposures with serum sodium, potassium, and chloride among infants with bronchopulmonary dysplasia

 

Change in serum Na, marginal means (mmol/L)

p-value

Change in serum K, marginal means (mmol/L)

p-value

Change in serum Cl, marginal means (mmol/L)

p-value

Novel Exposure

 Furosemide

−1.1

Ref.

0.0

Ref.

−3.4

Ref.

 Chlorothiazide

−0.9

0.813

−0.7

0.080

−2.8

0.567

PMA at Exposure

 36–40 weeks

−1.2

Ref.

0.0

Ref.

−3.6

Ref.

 40–44 weeks

−1.5

0.511

−0.1

0.663

−3.5

0.858

 44–48 weeks

−1.1

0.888

−0.5

0.107

−3.7

0.924

 >48 weeks

−0.6

0.431

0.0

0.829

−2.5

0.148

Diuretic exposure to follow-up lab time differencea

0.02

0.003

0.003

0.085

0.04

< 0.001

Diuretic doses between initiation and follow-up lab timea

−0.443

0.026

0.049

0.360

−0.746

0.004

Diuretic Doseb

 Low

−0.5

Ref.

0.0

Ref.

−2.7

Ref.

 High

−2.5

< 0.001

−0.4

0.285

−4.6

0.021

Electrolyte-Specific Covariates

No Increase in NaCl

−1.1

Ref.

    

Increase in NaClc

−0.3

0.418

    

Diuretic Route

      

 Intravenous

  

−0.1

Ref.

  

 Gastric

  

−0.2

0.932

  

 Post-Pyloric

  

−0.2

0.505

  

No Dexamethasone

    

−3.2

Ref.

Dexamethasone

    

−5.2

0.306

Male Sex

    

−2.8

Ref.

Female Sex

    

−3.6

0.290

  1. Na Sodium, K Potassium, Cl Chloride.
  2. Table depicts marginal means for change in serum Na, K, and Cl between pre- and post-lab values before and after novel diuretic administration estimated with a multivariable regression model using cluster-robust variance estimates to account for repeated observations within subjects. PMA at exposure, diuretic exposure to follow-up lab time difference, and dose were included a priori. Increase in NaCl supplementation during the exposure window between the pre- and post-lab times for sodium, concurrent dexamethasone use and sex for chloride, and medication route for potassium were included following association with novel furosemide and chlorothiazide exposure at p < 0.10 in bivariable analysis.
  3. aContinuous variable displayed as beta-coefficient instead of marginal mean.
  4. bDose for furosemide and chlorothiazide, low denotes ≤1 mg/kg/24h for furosemide and ≤10 mg/kg/24h for chlorothiazide, high denotes >1 mg/kg/24h for furosemide and >10 mg/kg/24h for chlorothiazide.
  5. cDenotes increase in enteral or parenteral NaCl supplementation between diuretic exposure and post-lab collection.