Early Postnatal Weight Loss: Is It a Problem?

In this issue of Neonatology, Kothapally et al. [1] report a systematic review and meta-analysis of published observational studies exploring the association between early postnatal weight loss (PWL) and outcomes in very preterm infants (<32 weeks). PWL in newborn preterm infants is common, highly variable, and often occurs without a known cause [2]. Overall, there are many causes for PWL, either excessive (EWL) or inadequate (IWL), which are well known. Less considered, uncertain, and often less documented are the implications of different weight-loss amounts. Generally, PWL has been attributed to reduced body water balance, the result of less parenteral and enteral fluid intake with higher water losses. Higher water losses come from relatively high urinary flow rates and evaporative losses. Evaporative water losses are higher in preterm infants due to their very thin skin, augmented by dry air and high air volume flows in NICUs, high air flows through incubators, and phototherapy. Evaporative water losses are variably limited by wrapping, humidification of incubators, humidification of inhaled oxygen or oxygen/air mixtures with CPAP and ventilation, and plastic drapes over infants under radiant warmers. These are variably offset by higher and highly variable intravenous fluid administration rates, particularly during resuscitations in the NICU and surgery, that are augmented by early and rapidly advanced enteral feeds. Urinary flow is high in normal infants but reduced in preterm infants with respiratory distress syndrome (RDS), sometimes for several days, with subsequent diuresis often a sign of resolving RDS. Prolonged insufficient nutrition can also limit weight gain, contributing to PWL.The implications of PWL are uncertain. Diuresis of excess fetal water content is considered essential for concentrating solutes in the kidney to develop renal function [3]. This varies by fluid administration rates along with solutes, especially sodium. High fluid intakes with earlier and higher intakes of sodium can lead to waterlogged lungs, aggravating RDS and PDAs [4]. Renal insults of many kinds lead to reduced renal blood flow and urine flow rates, potentially augmenting already higher water balances [5]. Variable use of diuretics, including caffeine, contributes to higher and variable renal water losses. Vasoactive drugs such as indomethacin and ibuprofen, and hormones such as vasopressin and catecholamines, affect blood pressure and renal perfusion. Depending on their mix and actions, they can both increase and decrease urinary water losses from the kidney.To address the causes and associations of PWL with adverse outcomes, Kothapally et al. [1] conducted their systematic review to synthesize the totality of evidence in the field, including both observational studies and reports from randomized controlled trials. They found that EWL (>15% from birth weight) and IWL (<5% from birth weight) had unique associations with certain morbidities. EWL was specifically associated with chronic lung disease (CLD) but also severe IVH, NEC, mortality, and mortality with IVH. IWL was specifically associated with CLD and mortality. The certainty of the evidence for these associations, however, was “low” or “very-low,” and very few publications met criteria for the meta-analysis – only 3, 3, 3, 3, and 2 studies were included in the mortality, IVH, CLD, PDA, and NEC categories, respectively. It is also unclear whether there were graded associations between EWL and IWL or whether they were only observed at extremes of weight loss and even weight gain. The authors did note, however, that there is some evidence that intermediate degrees of PWL are associated with better outcomes.The authors were not able to determine whether acute or longer term weight loss patterns had different or similar associations with adverse clinical conditions, although it is reasonable that inadequate nutrition would not contribute to short-term weight loss. It also is not clear how one would differentiate between the impacts of weight loss and adverse outcomes of the disorders associated with weight loss, EWL or IWL. The authors did try to adjust for outcomes associated with vasopressor use, disease severity scores, steroid treatment, fluid and energy intakes, gestational age, birth weight, PDAs, and growth status. Given the small number of studies and their limited information, however, it is difficult to assume that these adjustments might help differentiate significantly between adverse outcomes directly related to the disease and those resulting from pathophysiological instability associated with EWL or IWL. The authors correctly state that “Future research should focus on reporting adjusted outcomes that account for both disease severity and treatments that contribute to excessive or inadequate weight loss.”A few associations noted by Kothapally et al. [1] are reasonable. The common association of mortality among infants with EWL and IWL likely reflects that very preterm infants have significant morbidities with high risk of death. This group also has much more irregular feeding and diets that, combined with preterm birth and variable treatments for pathological conditions, put them at higher risk of NEC. High water and sodium intake with IWL has been associated with worse RDS and PDA pulmonary pathology [6], especially among infants who had variable blood pressures and received sodium chloride infusions [7, 8]. Longer term infants with CLD often receive diuretics to reduce lung water but then also receive sodium to replenish salt loss from the diuretics as well as to promote growth [9, 10]. More unstable, sicker infants often receive insufficient nutrition, especially enteral nutrition, which limits anabolism and weight gain from new cells and tissue growth. Unfortunately, Kothapally et al. [1] could not determine causes of either EWL or IWL, as data for the cofactors of severity of associated diseases, insufficient nutrition, and treatments could not be assessed.The results of the systemetic review and meta-analysis by Kothapally et al. [1] of studies that reported PWL in preterm infants demonstrate a paucity of evidence for many of the pre-specified pathological clinical outcomes, as well as heterogeneity of actual data and associations and limited certainty of evidence. The review does demonstrate statsitical significance of the relationship between EWL and IWL and several clinically relevant outcomes. This is useful information, even if with very low to low certainty, that might help guide future research in this area, which clearly deserves more attention [11].The author was a member of the journal’s Editorial Board at the time of submission.This commentary was not supported by any sponsor or funder.W.W. Hay Jr. is the sole author of this Commentary.