Oral-first rehydration for moderate neonatal hypernatremic dehydration: A protocol-driven advance with lessons for practice

Yaprak and Saygılı Karagöl, in their prospective study published in this journal, present a structured hypernatremic dehydration treatment (HDT) clinic protocol that challenges the prevailing intravenous-first approach to moderate neonatal hypernatremic dehydration (NHD) in term and near-term newborns. ¹ NHD, defined by a serum sodium exceeding 145 mEq/L in the setting of inadequate fluid intake, carries neurological consequences attributable not only to the degree of hyperosmolarity but equally to the speed and route of correction. The osmotic adaptation of neonatal brain tissue to a sustained hyperosmolar state, through intracellular accumulation of idiogenic osmoles, means that overly rapid sodium correction risks cerebral oedema and may worsen outcome more severely than the dehydration itself. ² No consensus treatment guidelines exist, but most experts recommend a target sodium reduction rate not exceeding 0.5 mEq/L per hour with correction completed over 48 h. ²

The authors evaluate this prospective HDT clinic protocol against a retrospective cohort managed before its implementation. The protocol’s defining feature is a stepwise rehydration hierarchy: oral feeding is the first-line modality, followed by orogastric tube administration on failure, with intravenous fluids reserved for predefined clinical indications including enteral failure, absence of urine output within 6 h, haemodynamic compromise, metabolic acidosis, or a serum creatinine exceeding 1.5 mg/dL. Intravenous fluid sodium content was tiered according to admission sodium level: 1/4 normal saline (34 mEq/L) for sodium up to 155 mEq/L, 1/3 normal saline (51 mEq/L) for 156–164 mEq/L, and 1/2 normal saline (77 mEq/L) for 165–169 mEq/L. Notably, no patient in Group 1 required the highest sodium tier, reflecting the effectiveness of the protocol in preventing escalation to severe presentations. The results are compelling: pure oral rehydration was achieved in 67.3% of the prospective group versus 25.9% of controls (P < 0.001), intravenous fluid use fell from 74.1% to 16.3% (P < 0.001), and the median sodium correction rate improved significantly from 1.05 to 0.66 mEq/L per hour (P = 0.001), with no seizures or mortality recorded in either group. The difference in correction rate between patients who received intravenous rehydration and those who did not was also significant, at 1.1 versus 0.66 mEq/L per hour, respectively (P < 0.001), further reinforcing the physiological rationale for prioritising the enteral route in clinically stable neonates.

The most clinically actionable finding in this study is the demonstration that serum sodium level alone is an insufficient determinant of rehydration route. In the multivariate analysis, group membership, that is, whether the infant was managed under the structured protocol, was the strongest independent predictor of intravenous fluid use in model 4 (OR 28.267, 95% CI 10.321–53.69, P < 0.001), far outweighing biochemical parameters. This finding quantifies what experienced neonatologists have long applied empirically: a haemodynamically stable neonate with a sodium of 160 mEq/L who accepts feeding and maintains urine output occupies a fundamentally different risk tier from one of identical biochemistry presenting with oliguria and metabolic acidosis. Protocol-guided clinical stratification, rather than sodium-level thresholds alone, is the operative variable determining both the safety and appropriateness of rehydration route.

An additional finding that deserves independent attention is the identification of serum uric acid as an independent predictor of intravenous rehydration need in model 2 (OR 1.495, 95% CI 1.092–2.00, P = 0.012). Hyperuricaemia in the setting of dehydration reflects reduced renal urate clearance and increased purine nucleotide catabolism under conditions of cellular hypoxia and volume depletion, a relationship established in dehydrated paediatric patients in whom serum uric acid correlated with blood urea nitrogen and was preferentially elevated in those with hypernatraemia. ³ In neonates, serum uric acid rises with perinatal hypoxic stress and is sensitive to the degree of prerenal insult. ⁴ Its emergence as a predictor of intravenous rehydration need in the present cohort suggests that it may function as a composite marker of metabolic severity beyond what serum sodium or creatinine alone conveys, and warrants prospective validation as a triage biomarker in moderate NHD.

Despite these strengths, several considerations temper immediate widespread adoption of this protocol. The sodium correction rate achieved in the prospective group, while significantly better than controls, remained at a median of 0.66 mEq/L per hour, exceeding the recommended safety threshold of 0.5 mEq/L per hour. Bolat et al., in a retrospective analysis of 81 neonates with hypernatremic dehydration, identified a correction rate above this threshold as an independent risk factor for mortality and neurological deterioration. ⁵ The authors acknowledge this gap and appropriately call for randomised controlled trials testing different oral rehydration volumes in moderately dehydrated cases. The absence of seizures in the current study is reassuring but should not be interpreted as evidence that 0.66 mEq/L per hour is universally safe: the study is underpowered to detect low-frequency neurological events, and the correction rate observed indicates that volume titration in oral rehydration remains imprecise. Refinement of feeding volumes per kilogram per hour across sodium strata within the moderate range represents a clear priority for future protocol iterations.

A second concern relates to the absence of standardised neurological assessment at discharge. The study records seizure-freedom as its primary safety endpoint, but formal neurological examination scoring and cranial ultrasound data are not reported for either group. Subclinical cerebral injury, including cortical signal change or evolving haemorrhage, may occur in the absence of overt seizures, particularly at correction rates that marginally exceed the safety threshold. ⁵ Future studies employing this protocol should incorporate cranial ultrasound and structured neurological assessment at discharge as predefined safety co-endpoints.

Third, the transition criteria for progressing from oral to orogastric feeding, and from orogastric to intravenous fluids, are described in clinical terms but the permissible time window for each step is not explicitly stated. This ambiguity limits protocol reproducibility across centres with varying nursing ratios and monitoring capacities. A time-anchored escalation matrix specifying the maximum duration at each rehydration tier before mandatory reassessment would make the protocol operationally transferable and reduce the risk of delayed escalation in less experienced settings.

Finally, although the protocol addresses rehydration route with commendable structure, it does not describe a concurrent standardised lactation support component. Since NHD arises almost exclusively from inadequate breastfeeding, and its recurrence risk is directly modifiable through lactation counselling initiated at diagnosis, the absence of a documented lactation intervention leaves the preventive dimension of management unaddressed. The authors note that the higher proportion of multiparous mothers in this cohort compared to Arora et al. (52% vs 67.6% primiparous) may reflect high caesarean section rates delaying lactogenesis in their setting, a contextual factor that further underscores the need for population-specific lactation strategies integrated within such protocols. ¹

Prior evidence comparing oral and intravenous rehydration in neonatal NHD has been limited. Erdemir et al., in a retrospective analysis of 75 neonates, demonstrated that sodium correction was significantly safer in the oral rehydration group at both 12 and 24 h, with one case of convulsion attributable to cerebral oedema occurring in the intravenous group. ⁶ The present study builds on this precedent by providing a prospectively evaluated, protocol-driven framework with predefined escalation criteria, tiered intravenous fluid sodium content, and 4-hourly electrolyte monitoring for higher sodium strata, representing a substantial improvement in methodological rigour over earlier descriptive comparisons.

In summary, Yaprak and Saygılı Karagöl contribute a well-structured, prospectively evaluated protocol that reduces unnecessary intravenous rehydration in moderate NHD and achieves a meaningfully safer sodium correction rate without short-term adverse events. The identification of serum uric acid as a triage predictor is a novel and biologically coherent finding. The limitations identified, namely, a correction rate exceeding the recommended threshold, absence of neurosurveillance endpoints, undefined step-transition time windows, and an absent lactation support component, are genuine gaps that future studies should address. Long-term neurodevelopmental follow-up of cohorts managed under oral-first protocols remains an unmet need. ⁷ Adequately powered, prospective, randomised trials with predefined neurological and developmental endpoints represent the necessary next step before this approach is adopted as standard of care for moderate NHD.