INTRODUCTION
Heart failure in neonates is a serious condition that contributes to significant morbidity and mortality worldwide. Effective management of fluid overload and maintaining hemodynamic stability are critical aspects of care in this population. Loop diuretics, particularly furosemide, are commonly used to reduce pulmonary congestion and improve cardiac function by promoting diuresis.[123] However, the optimal method of furosemide administration – continuous infusion versus intermittent bolus dosing – remains controversial, especially in neonates, where delicate fluid and electrolyte balances must be maintained.[4]
Previous pediatric studies have demonstrated the benefits of continuous infusion in critically ill infants after cardiac surgery,[5] while other research found comparable outcomes with intermittent bolus dosing.[67] Continuous infusion may offer more stable plasma drug levels and reduce fluctuations in intravascular volume, potentially leading to fewer side effects and better clinical outcomes.[89]
To date, no adequately powered neonatal randomized controlled trial (RCT) has directly compared continuous versus bolus furosemide administration for the treatment of heart failure. Given the high vulnerability of neonates to electrolyte disturbances, renal impairment, and prolonged ventilation, establishing the optimal approach is clinically important. This trial was designed to address this gap.
MATERIALS AND METHODS
Study design and participants
We performed a single-center RCT in the neonatal intensive care unit (NICU). Eligible subjects were term neonates diagnosed with congestive heart failure (e.g., from congenital heart defects or cardiomyopathy) requiring intravenous (IV) diuretics. Exclusion criteria included severe renal dysfunction (requiring dialysis), major electrolyte disturbances, hemodynamic instability needing emergency surgery, severe infection, or parental withdrawal.
A total of 60 neonates were enrolled and randomized 1:1; after protocol deviations and dropouts (e.g., early discharge), 52 infants completed the study (26 in each arm). The sample size (n = 26 per group) was determined based on prior estimates to achieve ~80% power for detecting clinically meaningful differences.
Interventions
In the continuous-infusion group, furosemide was given at 0.1 mg/kg/h IV, titrated up to 0.4 mg/kg/hour as needed. In the bolus group, furosemide 1 mg/kg IV was given every 12 h (with dose adjustments based on response). Other care (inotropes and ventilation) was at the clinicians’ discretion but applied equally.
Neonates were monitored daily for vital signs, FiO2, SpO2, urine output (UOP), blood pressure, heart rate, arterial blood gases, serum electrolytes (Na+, K+, HCO3−), renal function (blood urea nitrogen and creatinine), and echocardiographic ventricular function. Primary outcomes included pulmonary status (oxygenation and ventilator duration), renal function, and ICU length of stay; secondary outcomes included blood pressure and electrolyte stability.
Blinding
Although clinical staff could not be blinded, outcome assessors (laboratory personnel and echocardiographers) were blinded to the group assignment.
Statistical analysis
Data were analyzed using SPSS v. 26 (IBM Corp., Armonk, NY, USA). Normality was assessed using the Shapiro–Wilk test.
For normally distributed continuous variables, we compared groups using two-sided independent t-tests; for non-normal data, the Mann–Whitney U-test was employed. Categorical data were compared by χ² or Fisher’s exact test. Within-group (pre–post) changes were assessed by paired tests.
We used Bonferroni correction for multiple comparisons (setting stricter α thresholds where appropriate). Two-tailed P < 0.05 was considered significant. Continuous results are presented as mean ± standard deviation or median (interquartile range).
All analyses were intent-to-treat.
Ethical considerations
This clinical trial was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki. Ethical approval was obtained from the Ethics Committee of Kerman University of Medical Sciences (approval ID: IR.KMU.REC.1401.062, dated April 6, 2022). The study was prospectively registered in the Iranian Registry of Clinical Trials (IRCT) under the registration number IRCT20210317050737N1 on November 29, 2022.
Written informed consent was obtained from all parents or legal guardians of the neonates enrolled in the study prior to participation. All data were collected confidentially and used solely for research purposes. The study was carried out at Afzalipour Hospital, affiliated with Kerman University of Medical Sciences, Iran.
Study flow and participant enrollment
A total of 52 term neonates were assessed for eligibility and enrolled in the study. The flow of enrollment, allocation, follow-up, and analysis is shown in the CONSORT [Figure 1].
RESULTS
Baseline characteristics
The two groups are similar at baseline in age, weight, gender, and severity of heart failure. No significant differences were observed in initial pH, PaCO2, or creatinine levels between groups (all P > 0.05), indicating effective randomization [Table 1].
Pulmonary and clinical outcomes
After treatment, the continuous-infusion group showed significantly better gas exchange and faster recovery. Mean arterial pH was higher (7.34 ± 0.05 vs. 7.26 ± 0.06 in bolus, P = 0.011) and PaCO2 lower (46 ± 3 vs.
53 ± 4 mmHg, P = 0.007) in the infusion group, indicating improved ventilation. Consistent with this, infants on continuous infusion were weaned from mechanical ventilation sooner (mean 4.0 ± 1.2 vs. 5.5 ± 1.5 days, P = 0.009) and had shorter NICU stays (11.5 ± 2.8 vs.
15.2 ± 4.1 days, P < 0.001) [Table 1].
Within-group comparisons also revealed improvements in arterial pH, PaCO2, PO2, and bicarbonate after treatment, particularly in the continuous group [Table 2]. There was no difference in final FiO2 requirements; however, improved acid–base status and faster recovery suggest a clinical benefit.
Renal function and electrolytes
Both groups showed reductions in serum creatinine from baseline (continuous: 0.885 mg/dL to 0.777 mg/dL, P = 0.028; intermittent: 0.915 mg/dL to 0.823 mg/dL, P = 0.046), indicating improved renal perfusion with diuresis [Table 2].
However, at the study end, creatinine did not differ significantly between groups (P = 0.319), nor did blood urea or potassium levels [Table 1]. Sodium levels fell modestly only in the bolus group (mean 137.6→133.8 mEq/L, P = 0.001), though final serum Na+ was not significantly different between groups. Bicarbonate increased in both groups, reflecting improved metabolic balance.
Overall, electrolyte changes were mild and comparable, suggesting neither method caused severe electrolyte derangements.
Hemodynamics
Systolic blood pressure and heart rate remained similar between groups (P > 0.05), whereas diastolic blood pressure was significantly lower in the infusion group (29 ± 3 vs. 39 ± 5 mmHg, P = 0.001), likely reflecting more effective afterload reduction [Table 1]. No neonate developed severe hypotension or arrhythmias.
Inotrope use was comparable.
Echocardiography
Mean ejection fraction (EF) improved in both groups, with a greater but nonsignificant improvement in the continuous group at endpoint (46.9% vs. 46.35%, P = 0.913) [Table 1]. However, pre- and post-treatment analysis within each group showed a statistically significant rise in EF values [Table 2].
In summary, continuous infusion of furosemide was associated with significantly better respiratory outcomes (higher pH, lower PaCO2, shorter ventilation duration, and NICU stay) while maintaining renal and electrolyte stability [Figure 2]. Both methods effectively increased UOP and improved renal function from baseline, with no major safety differences.
DISCUSSION
This study investigated the comparative effects of continuous infusion versus intermittent bolus administration of furosemide on pulmonary and hemodynamic parameters in neonates with heart failure. Our findings demonstrated that continuous infusion significantly improved several critical clinical indicators, including a reduction in serum creatinine (from 0.885 to 0.7769 mg/dL, P = 0.028), a decrease in PCO2 (from 48.46 to 44 mmHg, P = 0.053), an increase in bicarbonate levels (HCO3) (from 19.75 to 22.42 mmol/L, P = 0.008), and an increase in oxygen partial pressure (PO2) (from 52.08 to 61.00 mmHg, P = 0.008). These changes reflect improved fluid balance, renal function, and respiratory efficiency in neonates receiving continuous infusion compared to those receiving intermittent bolus doses.[10]
While both administration methods effectively enhanced EF, the difference between the groups was not statistically significant, indicating that either method can be beneficial for improving cardiac function, although continuous infusion may provide a more stable clinical course. Therefore, the choice of administration should consider patient-specific clinical goals such as the urgency of diuresis and the risk of electrolyte imbalance.
Notably, a significant gender difference was detected between groups, suggesting a possible sex-based variation in response to furosemide therapy. Previous literature supports this, with Legato et al. and Madla et al. emphasizing sex-related differences in drug pharmacokinetics and efficacy.[1112] Conversely, Klein and Flanagan (2016) reported fewer gender-specific variations in cardiac treatments, indicating the necessity for further nuanced studies to understand these dynamics.[13]
Key clinical advantages of continuous infusion included significantly shorter hospital stays (11.46 vs. 15.23 days, P < 0.001) and faster ventilator weaning (4.00 vs. 5.46 days, P = 0.009), indicating quicker recovery.
Lower PCO2 levels in the continuous group (P = 0.007) also support enhanced pulmonary function. Interestingly, diastolic blood pressure was higher in the intermittent group (39.46 vs. 29.15 mmHg, P = 0.001), which might suggest improved hemodynamic stability in this cohort.
Other parameters such as sodium, potassium, urea, creatinine, systolic blood pressure, weight, and age showed no significant differences between groups (P > 0.05). Heart rate was higher in the continuous infusion group, but this difference did not reach statistical significance (P = 0.1).
Renal function improved in both groups, evidenced by decreased serum creatinine levels (continuous: P = 0.028; intermittent: P = 0.046), consistent with the nephroprotective effects of furosemide described by Kumar et al. and Chawla.[1415] Acid base status also improved, with significant increases in pH and bicarbonate levels, especially in the continuous infusion group. These results align with Zheng et al. (2021), who found that continuous infusion of furosemide in heart failure patients with moderate chronic renal dysfunction produced greater diuresis and improved renal outcomes compared with intermittent bolus[16] and Maeda et al., confirming that both administration methods contribute positively to acid–base homeostasis.[17]
Respiratory outcomes, including increased PO2 and decreased PCO2, improved in both groups, with the continuous infusion method showing a greater benefit. These findings support earlier studies by Tenzing et al. and Kjellberg, which demonstrated enhanced oxygenation and ventilation efficiency with both infusion protocols in neonates.[1819]
Hemodynamic parameters remained generally stable in both groups without significant changes over time, corroborating Landoni et al., who noted minimal impact of furosemide on blood pressure when appropriately dosed.[20] Neonates in the continuous group also benefited from shorter NICU stays, consistent with findings by Amer et al.[21]
Limitations
This study’s sample size was modest (26 infants per arm). Although we performed sample-size calculations for key outcomes, the power to detect small differences was limited. Thus, negative findings (e.g., no difference in creatinine or UOP) should be interpreted with caution; larger trials may reveal subtle effects.
We accounted for multiple comparisons using the Bonferroni correction; however, the residual risk of type I error remains. Being a single-center study, generalizability may be limited: our population had diverse etiologies of heart failure, and supportive care protocols may differ elsewhere. We did not blind caregivers, which could introduce bias in care decisions (though objective outcomes mitigate this).
Finally, follow-up was limited to the NICU stay; we do not know if one method affects longer-term renal or growth outcomes.
Future research should include larger multicenter RCTs and longer follow-up. It would be valuable to explore continuous infusion regimens (e.g., higher initial rates or combined diuretics) and to identify which subgroups (e.g., prematurity, renal insufficiency, and specific heart lesions) benefit most.
CONCLUSIONS
Continuous IV furosemide infusion and intermittent bolus dosing are both effective in neonatal heart failure, but with different strengths. In this trial, continuous infusion resulted in more stable diuresis, superior respiratory recovery (higher pH, lower PaCO2), and a shorter NICU stay, without compromising renal function. Intermittent boluses achieved similar renal and electrolyte outcomes but with slower pulmonary improvement.
These findings suggest that continuous infusion may be more appropriate for neonates with severe pulmonary edema or difficult ventilation, whereas bolus dosing is reasonable in less acute situations or when continuous infusion resources are limited. Ultimately, therapy should be tailored to the individual patient, balancing the need for rapid decongestion against the ease of administration and the risk of electrolyte shifts.
Conflicts of interest
There are no conflicts of interest.