Fluid restriction after pediatric cardiac surgery is independently associated With shorter mechanical ventilation: a retrospective cohort study With adjustment for cardiopulmonary bypass time.

BACKGROUND

Cardiopulmonary bypass (CPB) for pediatric cardiac surgery triggers a systemic inflammatory response driving postoperative fluid retention, prolonged ventilation, and acute kidney injury (AKI). The Starship Hospital guideline provides a structured framework for postoperative fluid restriction, but North American observational validation adjusting for CPB time has been lacking.

OBJECTIVE

To evaluate the association between a Starship-derived fluid restriction protocol and clinical outcomes, with adjustment for operative complexity.

METHODS

Retrospective single-center cohort of 329 pediatric patients undergoing structural CHD surgery with cardiopulmonary bypass (April 2017-February 2025): 259 pre-protocol, 70 post-protocol. Non-CPB lateral thoracotomy cases were excluded a priori. Protocol parameters were embedded in the mandatory order set, substantially standardizing fluid prescribing. Outcomes were analyzed with linear and generalized linear models (gamma; negative binomial for ventilation days), adjusted for age, weight, and STAT category, with prespecified sensitivity analyses adding CPB time, post-CPB TEE ventricular function, and procedure year.

RESULTS

Mean CPB time was shorter post-protocol (83.0 vs. 101.2 min; p = 0.008); modified ultrafiltration was not used. Adjusted for age, weight, and STAT, post-protocol status was associated with 49% lower expected ventilation days (IRR: 0.51, 95% CI: 0.37-0.69; p < 0.001), 34% lower LOS (ratio: 0.66, 0.50-0.88; p = 0.005), and 28% lower furosemide exposure (ratio: 0.72, 0.64-0.83; p < 0.001). After adjustment for CPB time, the ventilation effect persisted (IRR: 0.63; p = 0.004); adding post-CPB TEE ventricular function, the effect remained significant (IRR: 0.69, 0.50-0.96; p = 0.026). AKI did not differ unadjusted (42.5% vs. 37.1%, p = 0.49) or adjusted. Postoperative hypernatremia (Na >145) fell from 78.7% to 62.9% (adjusted OR: 0.44, 0.22-0.89; p = 0.022).

CONCLUSIONS

A Starship-derived fluid restriction protocol was independently associated with shorter mechanical ventilation, reduced furosemide exposure, and reduced hypernatremia, robust to full adjustment for operative covariates. AKI and LOS effects attenuated under full adjustment. Findings are hypothesis-generating and support proactive fluid restriction as a candidate intervention warranting prospective evaluation.