Early Chloride Intake Does Not Parallel That of Sodium in Extremely-Low-Birth-Weight Infants and May Impair Neonatal Outcomes

Kermorvant-Duchemin, Elsa*; Iacobelli, Silvia; Eleni-Dit-Trolli, Sergio*; Bonsante, Francesco; Kermorvant, Christopher*; Sarfati, Gilles; Gouyon, Jean-Bernard; Lapillonne, Alexandre*

Journal of Pediatric Gastroenterology & Nutrition: May 2012 - Volume 54 - Issue 5 - p 613–619
doi: 10.1097/MPG.0b013e318245b428
Original Articles: Hepatology and Nutrition

Background and Objective: Accurate data on the optimal chloride (Cl) intake in premature infants are scarce. The aim of the present study was to describe Cl intakes in the first 10 days of life and to assess the relations between high Cl intakes and corrected serum Cl level or markers of severe acidosis in infants <28 weeks’ gestation.

Methods: Retrospective cohort study including all of the infants <28 weeks admitted to the neonatal intensive care unit during a 3-year period and cared for from birth until day 10 or more.

Results: Fifty-six infants were included. Cumulative total Cl intakes reached 9.6 ± 3.7 mmol/kg at day 3 and 49.2 ± 13.5 mmol/kg at day 10. Inadvertent intakes (from intravenous fluids other than parenteral nutrition) represented on average 70% of total Cl intakes in the first 3 days. Difference between Cl and sodium intakes reached 7.8 ± 4.8 mmol/kg at day 10 and mainly originated from parenteral nutrition. By multivariate analysis, cumulative Cl intake >10 mmol/kg during the first 3 days was an independent risk factor of base excess <−10 mmol/L. Cumulative Cl intake >45 mmol/kg during the first 10 days was an independent risk factor of corrected chloremia >115 mmol/L and of base excess <−10 mmol/L.

Conclusions: Cumulative Cl intake >10 mmol/kg during the first 3 days (ie, 3.3 mmol · kg−1 · day−1 on average) and >45 mmol/kg during the first 10 days (ie, 4.5 mmol · kg−1 · day−1 on average) may have unwanted metabolic consequences and should be avoided. Imbalance between electrolytes provided by the parenteral nutrition solution need to be detected and corrected.

*AP-HP, Department of Neonatology, Necker-Saint-Vincent de Paul Hospital, Paris, the Paris Descartes University, Paris

Department of Paediatrics, Neonatal Intensive Care Unit, Dijon University Hospital, Dijon

AP-HP, Department of Biochemistry, Groupe Hospitalier Cochin-Saint-Vincent de Paul, Paris, France.

Address correspondence and reprint requests to Elsa Kermorvant, MD, Department of Pediatrics and Neonatal Intensive Care, Necker Hospital, 149 rue de Sevres, 75015 Paris, France (e-mail: elsa.kermorvant@nck.aphp.fr).

Supplemental digital content is available for this article. Direct URL citations appear in the printed text, and links to the digital files are provided in the HTML text of this article on the Journal's Web site (www.jpgn.org).

Received 8 October, 2010

Accepted 11 November, 2011

The authors report no conflicts of interest.

Article Outline

Chloride (Cl) is the most abundant anion in extracellular fluid, where it is involved in maintaining osmotic pressure, hydration, and ionic neutrality (1); however, data about optimal Cl intakes in premature infants are scarce, and present recommendations often rely on expert opinion (1). For most authors, the intake and output of Cl follow those of sodium (1,2). In clinical practice, Cl intakes of extremely-low-birth-weight (ELBW) infants are seldom controlled, and greater attention is often focused on sodium and potassium intakes; however, sources of Cl are numerous in perinatal care. Normal saline to maintain the patency of the indwelling arterial lines is used in many neonatal intensive care units (NICUs), and electrolytes, amino acids, and calcium in parenteral nutrition solutions can be provided as Cl salts. Therefore, the electrolyte balance (resulting from intakes of sodium, potassium, and Cl) can be different in infants fed parenterally compared with infants fed enterally.

High Cl load and hyperchloremia have been associated with metabolic acidosis in adults and children (3–5). Acidosis itself has been associated with adverse neurological outcomes in premature infants (6,7). Therefore, Cl intakes and electrolyte balance seem potentially important components of nutrition to evaluate. Our primary objective was to describe the Cl intakes in the first 10 postnatal days in premature infants <28 weeks’ gestation. Our second objective was to assess in this population the relations among high Cl intakes and serum Cl level, markers of severe metabolic acidosis, and neonatal outcomes.

Back to Top | Article Outline


Study Population

All of the extremely preterm infants <28 weeks’ gestation consecutively admitted to the NICU of Saint Vincent de Paul Hospital (tertiary level center) in Paris, France, from January 1, 2004 to December 31, 2006, and cared for from birth until at least postnatal day 10 were included. Babies who died within the first 10 postnatal days or those who were transferred to the NICU after day 1 were not included.

Back to Top | Article Outline

Outcome Measures

Cl intakes were calculated for total and cumulative intakes for each day of the first 10 postnatal days, and compared with sodium intakes. Total Cl intakes were defined as high when >10 mmol/kg in the first 3 days of life and >45 mmol/kg in the first 10 days of life. These cutoff values corresponded to the nearest round figure of the median total intakes for each period of time, and corresponded to an average intake of 3.3 mmol · kg−1 · day−1 for the first 3 days and 4.5 mmol · kg−1 · day−1 for the first 10 days.

Outcome measures to assess the effect of high Cl intakes were exposure to a corrected Cl (corrCl) level ≥115 mmol/L for 2 days or more; metabolic acidosis defined as minimum base excess (BE) <−10 mmol/L or minimum bicarbonates ≤12 mmol/L or minimum pH <7.2; and presence of intraventricular hemorrhages (IVH) of grade 2 or more. In the absence of a clearly established definition of acidosis and hyperchloremia (8), these cutoff values have been chosen to reflect severe hyperchloremia and severe metabolic acidosis. The duration of exposure to hyperchloremia (2 days) was chosen to avoid taking into account transient abnormalities.

Back to Top | Article Outline

Data Collection

Clinical Data

Clinical data were obtained from each infant's medical records. Documented data of pregnancy included number of fetuses, maternal hypertension or diabetes, and use of antenatal steroids. Antenatal steroid treatment was considered as complete if the first dose had been administered at least 48 hours before delivery. Documented clinical data of the newborn included gestational age, birth weight, Apgar and Clinical Risk Index for Babies scores (9,10), place of birth, administration of hydrocortisone hemisuccinate and/or fluid replacement for hypotension at day 1, administration of surfactant for a respiratory distress syndrome, treatment for a patent ductus arteriosus, sepsis, and IVH. Gestational age was determined by the first term ultrasonography. Percentile birth weight was defined using Fenton's charts (11). Small-for-gestational-age infants were defined as birth weight <10th percentile for gestational age. Early-onset sepsis was defined as sepsis occurring within 72 hours of age associated with a positive bacterial culture and/or clinical symptoms associated with infection (premature labor, fever >38°C, tainted amniotic fluid) in the mother. IVH were graded according to the method of Papile et al (12). The worst IVH was recorded for statistical analysis. All of the ultrasound scans were performed by pediatric radiologists.

Back to Top | Article Outline

Nutritional Data

According to parenteral nutrition practices in the unit, glucose-1-phosphate disodium tetrahydrate was used as a source of phosphate and sodium (up to 1.8 mmol · kg−1 · day−1 of phosphorus), sodium chloride as a source of sodium and chloride (up to 3 mmol · kg−1 · day−1 of Cl), and potassium chloride as a source of potassium and chloride (up to 4 mmol · kg−1 · day−1 of potassium and 4 mmol · kg−1 · day−1 of Cl). In addition, Primene (Baxter, Maurepas, France) as a source of amino acids, provided 0.2 mmol/g amino acids of Cl (up to 0.9 mmol · kg−1 · day−1 of Cl). Sodium acetate was not available.

Nutritional data (fluids, electrolytes, macronutrients) were obtained daily for the first 10 days after birth from the daily orders, and actual intakes were calculated using the nurse records. Total intakes were calculated by adding parenteral and enteral intakes. Cumulative intakes during the first 10 postnatal days were obtained by adding the daily enteral and parenteral intakes. Inadvertent intakes were defined as intakes from all intravenous fluids other than parenteral nutrition, such as those used for drug dilution and administration, or administered to maintain the patency of indwelling catheters (13). All were included in the calculation of total intravenous fluids and electrolytes intakes. Cl intakes from medications were based on French databases (UBM Medica, Issy les Moulineaux; Thériaque, CNHIM, Baillargues). Human milk was assumed to contain 1.6 mmol/100 mL Cl and 1.1 mmol/100 mL sodium (14,15). Formula intakes were based on published manufacturers’ figures. Infant daily weight was obtained using a calibrated infant scale to the nearest 5 g and the daily percentage of weight loss was determined.

Back to Top | Article Outline

Biological Data

All serum electrolytes, CO2, blood urea, creatinine, calcium, and phosphorus concentrations performed for each infant in the first 10 postnatal days were obtained from the electronic database of the hospital biochemistry laboratory. When serum electrolytes concentrations were performed several times per day, the testing with the highest value of Cl was used for statistical analyses. Capillary blood gas data (pH, PaCO2, PaO2, BE) and C-reactive protein were collected from each infant's chart. The frequency of laboratory tests was decided by the primary physician based upon the clinical status of the infants. Generally, infants had their serum electrolytes monitored every 24 hours in the first days of life. Blood gases were monitored every 2 to 4 hours as long as the infants were on mechanical ventilation. When several blood gases were obtained on the same day, the gases considered for statistical analyses were those with the median value for pH. For all of the analyses, serum Cl level was corrected (corrCl) to take into account variations in free water and natremia, according to the following formula (16):

corrCl = measured serum Cl/measured serum sodium × 140,

where 140 is the normal serum sodium level. Corrected chloremia thus reflects the variation in chloremia not caused by dehydration, which increases sodium and Cl concentrations proportionally.

Back to Top | Article Outline

Statistical Analyses

All of the data were consolidated into a MySQL database (Oracle, Redwood City, CA), allowing extraction of the data from different sources for statistical analyses. Statistical analyses were done using the Minitab 13.3 software (Minitab Inc, State College, PA). Daily Cl intakes were described using standard descriptive statistical methods (percentages, averages, standard deviation). The associations between cumulative Cl intakes and daily-corrected chloremia, pH, BE, and bicarbonates were evaluated by linear regression analysis. The effect of high Cl intakes (ie, 10 mmol/kg in the first 3 days and 45 mmol/kg in the first 10 days) on Cl plasma levels, acid-base parameters, and IVH were evaluated by univariate and multivariate analysis after discretization of continuous data (eg, gestational age, birth weight) to obtain 2 groups of similar size. Univariate analyses were performed with the χ2 test for the qualitative and discontinuous variables and by the unpaired t test for the continuous variables. Logistic regression models were developed to quantify the associations between Cl intakes in the first 3 or 10 days of life and corrCl plasma levels, markers of acidosis, and IVH, after adjustment for potentially confounding factors. P < 0.05 was considered significant.

Back to Top | Article Outline

Ethics Information

According to French law, neither ethical approval nor informed consent is required in noninterventional, retrospective cohort studies.

Back to Top | Article Outline


Study Population

Fifty-six neonates consecutively admitted in the NICU satisfied the inclusion criteria. Clinical characteristics of infants enrolled in the study are presented in Table 1. The mean (±standard deviation [SD]) gestational age and the mean (±SD) birth weight were 26.3 weeks (±0.8) and 885 g (±152), respectively. Data regarding the outcome parameters (corrCl plasma level, markers of acidosis, IVH) were available for all of the infants. All of the infants were tested for serum electrolytes (median number [IQ] of days with at least 1 laboratory test: 6 [5–7]); 23 infants (41.1%) presented with a corrected chloremia ≥115 mmol/L, and 18 (32.1%) with a corrected chloremia ≥115 mmol/L for ≥2 days. Mean corrected and uncorrected serum Cl levels per day are presented in the online-only supplementary Figure 1 (http://links.lww.com/MPG/A87), along with mean CO2 and BE. Markers of acidosis were present as follows: 27 (48.2%) infants had a minimum BE <−10 mmol/L, 24 (42.8%) had a minimum pH <7.2, and 21 (37.5%) had a minimum bicarbonate ≤12 mmol/L. Time-dependent serum levels of bicarbonates and BE during the first 10 days showed a progressive worsening of metabolic acidosis parameters during this period (supplementary Fig. 1, http://links.lww.com/MPG/A87). Fifty-one percent of the infants had a maximal serum sodium >150 mmol/L. Mean (±SD) water and protein intakes during the first 10 days were 113 mL · kg−1 · day−1 (±8.7) and 3.1 g · kg−1 · day−1 (±0.4), respectively. IVH of grade 2 or higher were observed in 27 (48.2 %) infants.

Back to Top | Article Outline

Cl Intakes

Cl intakes in the first 10 days of life are presented in Figure 1A. Cumulative total Cl intakes were (mean ± SD) 9.6 ± 3.7 mmol/kg at day 3 and 49.2 ± 13.5 mmol/kg at day 10.

Contribution of the 2 main sources of Cl (ie, parenteral nutrition and inadvertent intakes) to the total Cl intake is shown in Figure 1B. Inadvertent intakes were predominant in the first 3 days, representing 80.4%, 75.9%, and 45% of total Cl intakes at day 1, 2, and 3, respectively. They stabilized around 20% of daily total intakes beyond day 5. At day 3, cumulated inadvertent intakes and Cl from parenteral nutrition represented 71.7% and 27.2% of total cumulated Cl intake, respectively. At day 10, cumulated inadvertent intakes and Cl from parenteral nutrition represented 37.5% and 59.2% of total cumulated Cl intake, respectively. Contribution of medications was low among inadvertent Cl intakes (mean 0.07% of total intakes); Cl from sodium chloride represented the major part of the latter. Chloride intake from enteral intakes was low, reaching a maximum of 0.41 mmol/kg at day 10 (6.9% of total intakes).

Cl intakes did not strictly parallel sodium intakes (Fig. 1A–C). The difference between Cl and sodium intakes increased from birth to day 5 to stabilize at approximately 1 mmol · kg−1 · day−1. At day 10, the difference between cumulative Cl and sodium intakes reached (mean ± SD) 7.8 ± 4.8 mmol/kg. As shown in Figure 1B, this difference mostly resulted from parenteral nutrition intakes. On average, 72% of this difference was caused by an excess of Cl from amino acid solution and potassium chloride compared with sodium intake from glucose-1-phosphate disodium tetrahydrate. The difference between Cl intake and sodium intake from enteral nutrition was low (mean difference during the study period was 0.06 mmol · kg−1 · day−1).

Of the 56 infants, 26 (46.4%) received >10 mmol/kg of Cl in the first 3 postnatal days and 33 (58.9 %) received >45 mmol/kg in the first 10 postnatal days. Unadjusted comparisons between infants with low and high Cl intakes are presented in Table 1. Water intakes during the first 3 days were higher in infants who received >10 mmol/kg of Cl (mean ± SD) (248.8 ± 36.3 vs 216 ± 28.9 mL/kg/3 days, P = 0.004). Protein intakes (5.8 ± 1.6 vs 5.8 ± 1.6 g/kg/3 days, P = 0.91) and maximal serum sodium level (152.6 ± 7.2 vs 149.6 ± 4.9 mmol/L, P = 0.08) were not significantly different between the 2 groups. Water intakes during the first 10 days were slightly higher in infants who received >45 mmol/kg of Cl (mean ± SD) (1151 ± 92.5 vs 1102 ± 71.3 mL/kg/10 days, P = 0.04). Protein intakes (31.7 ± 3.7 vs 29.9 ± 4.3 g/kg/10 days, P = 0.11) and maximal serum sodium level (152.1 ± 7 vs 149.4 ± 4.6 mmol/L; P = 0.11) were not significantly different between the 2 groups.

Back to Top | Article Outline

Cl Intakes and Corrected Serum Cl Level

Linear regression analysis showed a significant association between corrected serum Cl and cumulative Cl intakes until blood measurement (R2 = 0.3047; P < 0.0001) (Fig. 2A). Serum Cl corrected for free water correlated better with Cl intakes than uncorrected serum Cl (Fig. 2A and B).

By univariate analysis, significant associations were documented between corrected serum Cl level >115 mmol/L for ≥2 days and Cl intake >45 mmol/kg in the first 10 days (unadjusted odds ratio [95% CI] 23.4 [2.8%–194.2%]), lower birth weight (P = 0.038), absence or incomplete antenatal steroid treatment (P = 0.036), respiratory distress syndrome (P = 0.046), treated patent ductus arteriosus (P = 0.02), maximal natremia >150 mmol/L (P = 0.007), maximal plasma protein >50 g/L (P = 0.036), and HIV II–IV (P = 0.013). Multivariate regression analysis adjusting for factors that were shown to be significant by univariate analysis indicated that cumulative Cl intakes >45 mmol/kg in the first 10 days remained significantly associated with a higher risk of a corrected serum Cl level >115 mmol/L for ≥2 days (Table 2). Maximal serum sodium level >150 mmol/L did not reach significance after adjustment for other risk factors.

Back to Top | Article Outline

Cl Intakes and Markers of Severe Acidosis

Linear regression analysis showed a significant correlation between cumulative Cl intakes and all of the acidosis parameters (BE, plasma bicarbonates, pH) measured daily. The strongest correlation was found with BE (R2 = 0.45; P < 0.0001) (online-only supplementary Fig. 2, http://links.lww.com/MPG/A87); results of the univariate analysis for this parameter only are therefore presented in Table 3. Separate models of multivariate regression analysis were developed to study the association between minimal BE <−10 mmol/L and either Cl intakes >10 mmol/kg or >45 mmol/kg during the first 3 and 10 days, respectively. After controlling for potential confounders, both cumulative Cl intakes >10 mmol/kg in the first 3 days and cumulative Cl intakes >45 mmol/kg in the first 10 days remained independent predictors of minimal BE <−10 mmol/L (Table 3).

Other models of multivariate regression analysis including either minimal pH <7.2 or minimal plasma bicarbonate ≤12 mmol/L instead of BE showed consistent results, Cl intake >45 mmol/kg in the first 10 days remaining an independent risk factor of minimal plasma bicarbonate ≤12 mmol/L, and Cl intake >10 mmol/kg in the first 3 days remaining an independent risk factor of minimal pH <7.2 (data not shown).

Treatment for hypotension was not significantly associated with acidosis parameters by univariate analysis except with minimal plasma bicarbonate ≤12 mmol/L (P = 0.032), but the association did not remain significant after adjustment for other risk factors (P = 0.882). Protein intakes and weight loss were not significantly associated with any of the acidosis parameters (data not shown).

Back to Top | Article Outline

Cl Intakes and IVH

High Cl intakes in the 3 and 10 first postnatal days both correlated with IVH of grade 2 or more by univariate analysis (P = 0.017 and P = 0.001, respectively), but did not reach significance after adjustment for confounding variables (P = 0.068 for Cl intakes >45 mmol/kg during the first 10 days). Hypernatremia >150 mmol/L was not significantly associated with IVH in our population, nor were markers of acidosis (data not shown).

Back to Top | Article Outline


Recommended Cl intakes in premature infants are variable, ranging from 0 to 5 mmol · kg−1 · day−1 in the first week of life (1,2,17). For most authors, Cl intakes follow those of sodium (1,2). In our study, Cl intakes were superior to sodium intakes, leading to a mean cumulative difference of nearly 8 mmol/kg during the first 10 days of life. In our study, the discrepancy between Cl and sodium intakes was mainly caused by parenteral nutrition, mostly from amino acid solution and potassium chloride. This inappropriate Na/Cl ratio in the composition of the local parenteral nutrition may represent a limitation of the study because composition of parenteral nutrition solution may be different in other NICUs. Furthermore, inadvertent intakes were important contributors of total intakes, representing on average nearly 70% of total Cl intakes in the first 3 postnatal days. These data underline the need to take into account all of the sources of Cl in the calculation of total intake and to detect and correct any imbalance in total parenteral electrolyte intake.

In the present study performed in extremely preterm infants, we showed that Cl intakes >10 mmol/kg (ie, 3.3 mmol · kg−1 · day−1 on average) and >45 mmol/kg (ie, 4.5 mmol · kg−1 · day−1 on average) in the first 3 and 10 postnatal days, respectively, were associated with a higher risk of hyperchloremia. Of note, the scatterplot from Figure 2A suggests that the relation between Cl intake and serum Cl may be more piecewise linear than fully linear and that serum Cl level increases less when cumulative Cl intake exceeds approximately 18 mmol/kg (1.8 mmol · kg−1 · day−1 on average).

High Cl intakes were also associated with metabolic acidosis. Accordingly, time-dependent serum levels of bicarbonates and BE during the first 10 days showed (without intervention) a progressive worsening of metabolic acidosis parameters during this period. These results are in agreement with previously published reports and suggest that the basic principles of Cl physiology are also valid for ELBW infants. The association between high Cl load from 0.9% saline for fluid resuscitation and metabolic acidosis has been described in animal models (18) and in critically ill adults and children (3–5). In premature infants, Groh-Wargo et al (19) reported that a total Cl load in excess of 6 mmol · kg−1 · day−1 was associated with a higher incidence of metabolic acidosis. These data suggest that in the presence of metabolic acidosis, a careful evaluation and adaptation of Cl intakes (notably inadvertent Cl intakes) and of parenteral electrolyte balance (Na, K, Cl) would be an important point to address among other therapeutic interventions.

Several other approaches to minimize nutrition-induced metabolic acidosis in preterm infants have been proposed in the literature. In preterm infants fed enterally, the increase in alkali (mainly potassium) content of preterm formulas affects renal regulation of acid-base balance and prevents the development of incipient late metabolic acidosis (20). The acid load of formulas is estimated by the difference between cations and anions and can be calculated using various equations (ie, alkali excess and potential renal acid load of food). This approach can be used to determine the acid load of parenteral nutrition to prevent the overload of the renal capacity for net acid excretion, which is known to be transiently decreased in preterm infants (20). The use of sodium acetate has also long been proposed to correct metabolic acidosis in premature infants (21–23). Effect of acetate is believed to result from a decrease in Cl intake, but also from its metabolism to bicarbonate (23). Nevertheless, concerns regarding hypercarbia with high acetate load (>6 mmol · kg−1 · day−1) have been raised (23).

The retrospective design of our study prevented us to include renal electrolyte excretion and bicarbonate urinary losses in the analysis of acidosis parameters; however, several studies suggest that urinary losses of bicarbonates do not significantly contribute to the metabolic acidosis occurring during the first week of life. Ramiro-Tolentino et al (24) showed that renal bicarbonate excretion of ELBW infants is low (13% of sodium loss; cumulative loss 2 mEq/kg) and the net bicarbonate balance is positive in the first 4 days of life. A study by Zilleruelo et al (25) suggested that the bicarbonate tubular reabsorption was adequate during metabolic and/or respiratory acidosis in preterm infants. Furthermore, in a study comparing 32 ELBW infants to 36 later preterm infants, Sato et al (26) found that bicarbonate excretion was higher in ELBW infants in the first 2 days of life but improved rapidly afterward.

The mechanisms of metabolic acidosis induced by a high Cl load have received renewed attention with the publication of Stewart's model of acid-base physiology (27,28). Using the conventional Henderson-Hasselbalch equation, the most commonly offered explanations for the mechanism of hyperchloremic acidosis following fluid resuscitation include renal failure to excrete hydrochloric acid, and dilution of plasma bicarbonate by intravenous saline (4,29). According to Stewart's approach, plasma bicarbonate and consequently blood pH are determined by 4 systems. The first one is the partial pressure of carbon dioxide. The second is the strong ion difference (SID), that is, the difference in concentration between strong anions and cations, the major ones in blood being Na, K, and Cl. The 2 other factors determining pH are the total concentration of weak acids (proteins, mainly albumin), and the presence, in pathological conditions, of others acids such as lactic acid or ketoacids. The SID is calculated as the charge difference between the sum of measured strong anions (Na, K, Ca, Mg) and measured strong anions (Cl, lactate). A decrease in SID results in a decrease in pH. Therefore, an increase in the plasma Cl relative to Na decreases the plasma SID and lowers the pH. Chloride concentration needs to be corrected for the changes in free water and natremia so that these do not constitute a bias when interpreting effects of Cl concentration on acid-base balance. Stewart's theory of acid-base physiology has been put into practice in numerous studies (3–5,29). The role of Cl load in acidosis was demonstrated in a prospective randomized crossover study in healthy volunteers after administration of colloid solutions. In this study, decreases in BE were observed after administration of a high Cl-containing solution but not after albumin (29). Gilfix et al (30) also showed in a clinical setting that a change in corrCl concentration was the major factor affecting acid-base status in patients from 2 ICUs and an emergency room. The effect on BE because of changes in corrCl concentration varied from −14.8 to 11.9 mmol/L, whereas the effect from changes in free water varied from −3.6 to 3 mmol/L (30). Our results, in a population of extremely preterm infants, are consistent with all of these observations.

By univariate analysis, we observed an association between early high Cl intakes and a greater risk of IVH, with a trend toward significance for high Cl intakes during the first 10 days after controlling for potential confounding factors by logistic regression (P = 0.068). Although high Cl intakes may merely reflect the severity of the initial illness, further studies are required to clarify this association. Hypernatremia and high sodium intakes have been shown to be associated with an increased of risk of IVH (31,32), but effect of Cl intakes was not examined in either study. Metabolic acidosis has been reported as a risk factor for IVH in several studies (6,7,33,34); although a lack of power cannot be excluded because of the small number of cases, it should be noted that none of the markers of acidosis reached statistical significance for IVH in our population. Long-term consequences of high Cl intakes on developmental outcome remain to be evaluated.

Back to Top | Article Outline


We have shown that cumulative Cl intakes >10 mmol/kg during the first 3 days (ie, 3.3 mmol · kg−1 · day−1 on average) and 45 mmol/kg during the first 10 days (ie 4.5 mmol · kg−1 · day−1 on average) were independently associated with undesirable metabolic events—hyperchloremia and metabolic acidosis. Our results suggest that the upper limit of recommended Cl intakes may be too high to prevent these conditions. In clinical practice, tight control of Cl intakes and close monitoring of electrolyte balance in parenteral nutrition in the first days of life of ELBW infants may represent important ways to lower the incidence of metabolic acidosis. The contribution of inadvertent intakes should be included in the calculation of total Cl intakes.

Back to Top | Article Outline


The authors are grateful to the ARFEN for its support.

Back to Top | Article Outline


1. Koletzko B, Goulet O, Hunt J, et al. Guidelines on Paediatric Parenteral Nutrition of the European Society of Paediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN) and the European Society for Clinical Nutrition and Metabolism (ESPEN), Supported by the European Society of Paediatric Research (ESPR). J Pediatr Gastroenterol Nutr 2005; 41 (suppl 2):S1–87.
2. Pointdexter B, Leitch C, Denne S. Nutrition and metabolism in the high-risk neonate—Pt 2—parenteral nutrition. In: Fanaroff A, Martin R, Walsh M, eds. Fanaroff and Martin's Neonatal-Perinatal Medicine: Diseases of the Fetus and Infant. 8th ed. St Louis: Elsevier Mosby; 2006:679–93.
3. Scheingraber S, Rehm M, Sehmisch C, et al. Rapid saline infusion produces hyperchloremic acidosis in patients undergoing gynecologic surgery. Anesthesiology 1999; 90:1265–1270.
4. Ho AM, Karmakar MK, Contardi LH, et al. Excessive use of normal saline in managing traumatized patients in shock: a preventable contributor to acidosis. J Trauma 2001; 51:173–177.
5. Skellett S, Mayer A, Durward A, et al. Chasing the base deficit: hyperchloraemic acidosis following 0.9% saline fluid resuscitation. Arch Dis Child 2000; 83:514–516.
6. Cooke RW. Factors associated with periventricular haemorrhage in very low birthweight infants. Arch Dis Child 1981; 56:425–431.
7. Levene MI, Fawer CL, Lamont RF. Risk factors in the development of intraventricular haemorrhage in the preterm neonate. Arch Dis Child 1982; 57:410–417.
8. Lawn CJ, Weir FJ, McGuire W. Base administration or fluid bolus for preventing morbidity and mortality in preterm infants with metabolic acidosis. Cochrane Database Syst Rev 2005:CD003215.
9. Parry G, Tucker J, Tarnow-Mordi W, et al. The CRIB (Clinical Risk Index for Babies) score: a tool for assessing initial neonatal risk and comparing performance of neonatal intensive care units. The International Neonatal Network. Lancet 1993;342:193–8.
10. Parry G, Tucker J, Tarnow-Mordi W. CRIB II: an update of the clinical risk index for babies score. Lancet 2003; 361:1789–1791.
11. Fenton TR. A new growth chart for preterm babies: Babson and Benda's chart updated with recent data and a new format. BMC Pediatr 2003; 3:13.
12. Papile LA, Burstein J, Burstein R, et al. Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weights less than 1,500 gm. J Pediatr 1978; 92:529–534.
13. Bathia J. Fluid and electrolyte management in the very low birth weight neonate. J Perinatol 2006; 26 (suppl. 1):S19–S21.
14. Schanler R, Atkinson S. Human milk. In: Tsang R, Uauy R, Koletzko B, et al, eds. Nutrition of the Preterm Infant. 2nd ed. Cincinnati: Digital Educational Publishing; 2005:333–56.
15. Neville M, McManaman J. Milk secretion and composition. In: Thureen P, Hay W, eds. Neonatal Nutrition and Metabolism. 2nd ed. Cambridge, UK: Cambridge University Press; 2006:377–89.
16. Balasubramanyan N, Havens PL, Hoffman GM. Unmeasured anions identified by the Fencl-Stewart method predict mortality better than base excess, anion gap, and lactate in patients in the pediatric intensive care unit. Crit Care Med 1999; 27:1577–1581.
17. Fuchs C, Jochum F. Water, sodium, potassium and chloride. In Tsang R, Uauy R, Koletzko B, et al, eds. Nutrition of the Preterm Infant. 2nd ed. Cincinnati: Digital Educational Publishing; 2005:201–44.
18. Kellum JA. Fluid resuscitation and hyperchloremic acidosis in experimental sepsis: improved short-term survival and acid-base balance with Hextend compared with saline. Crit Care Med 2002; 30:300–305.
19. Groh-Wargo S, Ciaccia A, Moore J. Neonatal metabolic acidosis: effect of chloride from normal saline flushes. JPEN J Parenter Enteral Nutr 1988; 12:159–161.
20. Kalhoff H, Manz F, Kiwull P, et al. Food mineral composition and acid-base balance in preterm infants. Eur J Nutr 2007; 46:188–195.
21. Ekblad H, Kero P, Takala J. Slow sodium acetate infusion in the correction of metabolic acidosis in premature infants. Am J Dis Child 1985; 139:708–710.
22. Richards CE, Drayton M, Jenkins H, et al. Effect of different chloride infusion rates on plasma base excess during neonatal parenteral nutrition. Acta Paediatr 1993; 82:678–682.
23. Peters O, Ryan S, Matthew L, et al. Randomised controlled trial of acetate in preterm neonates receiving parenteral nutrition. Arch Dis Child Fetal Neonatal Ed 1997; 77:F12–15.
24. Ramiro-Tolentino SB, Markarian K, Kleinman LI. Renal bicarbonate excretion in extremely low birth weight infants. Pediatrics 1996; 98:256–261.
25. Zilleruelo G, Sultan S, Bancalari E, et al. Renal bicarbonate handling in low birth weight infants during metabolic acidosis. Biol Neonate 1986; 49:132–139.
26. Sato T, Takahashi N, Komatsu Y, et al. Urinary acidification in extremely low birth weight infants. Early Hum Dev 2002; 70:15–24.
27. Stewart PA. Modern quantitative acid-base chemistry. Can J Physiol Pharmacol 1983; 61:1444–1461.
28. Sirker AA, Rhodes A, Grounds RM, et al. Acid-base physiology: the “traditional” and the “modern” approaches. Anaesthesia 2002; 57:348–356.
29. Waters JH, Bernstein CA. Dilutional acidosis following hetastarch or albumin in healthy volunteers. Anesthesiology 2000; 93:1184–1187.
30. Gilfix BM, Bique M, Magder S. A physical chemical approach to the analysis of acid-base balance in the clinical setting. J Crit Care 1993; 8:187–197.
31. Barnette AR, Myers BJ, Berg CS, et al. Sodium intake and intraventricular hemorrhage in the preterm infant. Ann Neurol 2010; 67:817–823.
32. Lim WH, Lien R, Chiang MC, et al. Hypernatremia and grade III/IV intraventricular hemorrhage among extremely low birth weight infants. J Perinatol 2011; 31:193–198.
33. Lavrijsen SW, Uiterwaal CS, Stigter RH, et al. Severe umbilical cord acidemia and neurological outcome in preterm and full-term neonates. Biol Neonate 2005; 88:27–34.
34. Ertan AK, Tanriverdi HA, Meier M, et al. Perinatal risk factors for neonatal intracerebral hemorrhage in preterm infants. Eur J Obstet Gynecol Reprod Biol 2006; 127:29–34.

acidosis; chloride intake; nutrition; premature infant

Supplemental Digital Content

Back to Top | Article Outline
Copyright 2012 by ESPGHAN and NASPGHAN