Keywords: Metabolic acidosis; Preterm infants; Sodium bicarbonate

Introduction

Metabolic acidosis with loss of bases is a common complication of prematurity. It is mainly due to renal immaturity [1]. Full performance of renal functions depends on gestational age and postnatal age, so that glomerular and tubular functions are immature [2,3]. Tubulopathy of preterm is responsible for excessive loss of bicarbonate due to a lack of tubular reabsorption of bicarbonate, electrolytes and small proteins that lead to metabolic acidosis, fluid, electrolyte disturbances, and weight problems [2,4,5]. Metabolic acidosis may have serious consequences: impaired cardiovascular function, cardiac arrhythmia, inflammation and decreased immune response [6]. Metabolic acidosis in prematurity has been associated with the development of intraventricular hemorrhage, periventricular leukomalacia and delayed psychomotor development [1]. For more than 60 years, sodium bicarbonate administration has been used in case of metabolic acidosis management in order to normalize the pH by acid-base reaction [7-10]. There are no recommendations in the treatment of metabolic acidosis of premature renal failure, although this treatment is widely used. Several authors, however, have described occasional severe complications such as intraventricular hemorrhage (IVH) following the use of bolus bicarbonate infusion in preterm newborns [1,11].

The main objective of our study was to evaluate the restriction of the indication of sodium bicarbonate in metabolic acidosis due to renal immaturity in preterm under 28 weeks of gestation during two periods and to assess the outcome of untreated patients during the second period.

The secondary objective was to evaluate the consequences of this change of practice on acid-base status.

Materials and Methods

Patients and methods: This is a retrospective monocentric study in the neonatal intensive care unit of the Hôpital Nord of Marseille (France) from two historical cohorts. All preterm infants under 28 weeks of gestation with metabolic acidosis were included. Metabolic acidosis was defined as pH less than 7.20 and EB less than or equal to -10 mmol / L. During the first period, from June 2011 to May 2012, patients were treated with continuous sodium bicarbonate infusion at 1 to 2 mEq / kg / day if the EB was less than or equal to -10 mmol / L, until correction of metabolic acidosis. During the second period, from October 2012 to September 2013, patients were treated if the EB was less than or equal to -13 mmol / L. The factors taken into account were: gestational age, birth weight, age at the time of treatment, pH, pCO2 and EB (before, during and at the end of sodium bicarbonate treatment), duration of treatment, the total dose administered, the occurrence of intraventricular hemorrhage stage III and / or IV according to the classification of Papile [12].

Primary outcome was the evolution of pH, pCO2 (mmHg) and EB (mmol/l). Secondary outcome were theoccurrence of IVH III and IV and duration of treatment. The first day (D 0) considered is the first day of treatment ie when the EB was less than or equal to - 10 mmol / L in the "first period" group and less than or equal to -13 mmol / l in the "second period" group. End day was the last day of treatment. For untreated patients, the D0 was when the patient had an EB less than or equal to -10 mmol / L and the end day was 3 days later.

Statistical analysis: The differences between the end of treatment and the day 0 were calculated, mismatched differences between the values were  ??observed between the end and the beginning of treatment for pH, PCO2 and EB. For untreated patients in the "second period", the difference between first and third day was calculated. The comparisons between the quantitative variables were made by Student's t-test, and by the χ2 test with a risk α fixed at 5%. The software used was PASW Statistics (V 17.0.2). The number of patients required was 32 patients (16 per group) as calculated with BioStat TGV (http://marne.u707.jussieu.fr/biostatgv/), with risk alpha 0.05 and beta 80%.

Ethics: The procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and regional) and with the Helsinki Declaration of 1964, as revised in 2013.

Results: Twenty-seven patients were included during the first period with an average gestational age of 26.5 (SD 2.5) weeks versus 38 patients during the second period with an average gestational age of 26.1 (SD 1.1) weeks (p = 0.41). The mean birth weight was 0.96 kg (standard deviation: 0.45) in the first period versus 0.90 kg (standard deviation: 0.17) in the second period (p = 0.57). The main results are summarized in (Table 1).