A Retrospective Study on Perioperative IV Fluid Restriction After a Disaster-Induced Fluid Shortage: Postoperative Pain and Complication Outcomes in Pediatric Adenotonsillectomies

Abstract

Intravenous fluids are integral to pediatric perioperative care, yet optimal fluid volumes during adenotonsillectomy remain debated. In 2024, Hurricane Helene disrupted IV fluid supply chains, necessitating an involuntary shift to restrictive intraoperative fluid administration. This event created an opportunity to evaluate whether reduced intraoperative fluids affected postoperative pain or complication rates in pediatric adenotonsillectomies. We conducted a retrospective cohort study of children under 12 years who underwent adenotonsillectomy between 1 October 2024, and 31 January 2025. Patients were stratified into a restrictive fluid group (<10 mL/kg) and a non-restrictive group (≥10 mL/kg). Collected data included demographics, intraoperative fluid volumes, postoperative FLACC pain scores, and documented complications. Pain scores were compared using Mann–Whitney U tests, due to non-normal data distribution with descriptive analysis. A total of 133 patients were included (63 restrictive, 70 non-restrictive). Mean postoperative FLACC scores were similar between groups (4.53 ± 2.62 vs. 4.57 ± 3.44; p = 0.50), with comparable operative times. Complications occurred in both groups without a consistent association with fluid strategy. These findings suggest that intraoperative fluid restriction below 10 mL/kg does not significantly affect postoperative pain or overall complication rates in pediatric adenotonsillectomy. Short-term fluid restriction may be safe in resource-limited settings, though prospective studies are warranted.

1. Introduction

In 2024, Hurricane Helene disrupted national IV fluid manufacturing and supply chains, necessitating an institutional shift toward restrictive intraoperative fluid administration [1]. This unplanned change in practice allowed for an opportunity for a retrospective comparison of pediatric patients managed under differing intraoperative fluid strategies within a uniform clinical setting. This was possible because a tertiary hospital initiated a protocol, which established a cut-off value that was verified internally by the institution. The nature of this protocol permitted some patients during this period to receive intraoperative IV fluids above and below this cut-off value, which created an opportunity to study the effects of IV fluid restriction in this patient population at this selected threshold. Such externally imposed practice changes offer a unique opportunity to evaluate real-world clinical outcomes without deliberate protocol modification.

Currently, fluid composition is a balanced isotonic crystalloid solution with 1–2% glucose, which is recommended for intraoperative maintenance in pediatric patients [2,3,4]. Acceptable balanced solutions include buffered Ringer’s, Plasma-Lyte, and Sterofundin, which are conventionally preferred over 0.9% saline to avoid hyperchloremic acidosis [5]. Recent findings suggest that there are no significant differences in adverse outcomes including hyperchloremic acidosis when using normal saline and lactated ringers perioperatively for fluid resuscitation [6]. The addition of 1–2% glucose is sufficient to prevent hypoglycemia, lipolysis, and ketosis without causing hyperglycemia [3,4].

When discussing infusion rates, intraoperative maintenance fluid rates should follow the Holliday-Segar “4-2-1” rule: 4 mL/kg/h for the first 10 kg, 2 mL/kg/h for the next 10 kg, and 1 mL/kg/h for each additional kilogram [3]. In practice, median infusion rates of approximately 4 mL/kg/h are safe and effective in infants undergoing surgery [2,3]. Moreover, for patients with a body weight exceeding 100 kg, fluid volumes should be calculated based on a maximal body weight of 100 kg [7].

Concerning too much fluid, excessive intraoperative IV fluid administration in pediatric surgical patients is associated with prolonged hospital stay, delayed recovery of bowel function, increased postoperative complications, and potential organ dysfunction [8,9,10]. There is also a concern with delayed gastrointestinal recovery as a prominent consequence of fluid overload. In children aged 6 months to 3 years undergoing elective open bowel surgery, those receiving conventional liberal fluid therapy experienced a significantly delayed recovery of bowel function, which can also result in severe nausea and emesis [3,10,11,12].

Adenotonsillectomy represents an ideal model for evaluating perioperative fluid strategies in pediatric surgery. It is among the most commonly performed pediatric otolaryngologic procedures, is typically short in duration, and it is associated with low rates of major perioperative complications [13]. Because patients are generally healthy, and procedures are standardized, adenotonsillectomy provides a relatively controlled clinical setting in which to examine the effects of intraoperative interventions, such as IV fluid administration, without the confounding influence of prolonged operative times or complex postoperative care pathways [12,13]. Despite its frequency, the evidence guiding optimal intraoperative fluid volumes for pediatric adenotonsillectomy remains limited, and most existing recommendations are extrapolated from adult surgery or higher-risk pediatric populations [3,4]. This gap is particularly relevant when perioperative practices must be adapted rapidly in response to external constraints, such as disaster-related resource shortages.

The objective of this study was to evaluate whether a disaster-induced reduction in intraoperative intravenous fluid administration was associated with differences in immediate postoperative pain or short-term complication rates in pediatric patients undergoing adenotonsillectomy. Leveraging a hurricane-related IV fluid shortage, we compared postoperative outcomes between patients managed with a restrictive intraoperative fluid strategy (<10 mL/kg) and those receiving non-restricted fluid volumes (≥10 mL/kg) within a single institutional setting.

This study aims to quantify the clinical perioperative effects of disaster-related intraoperative IV fluid restriction in a common pediatric surgical population. Postoperative pain scores and short-term complication rates are compared between fluid-restricted and non-restricted cohorts to determine whether moderate reductions in intraoperative fluid volume are associated with meaningful differences in early recovery after adenotonsillectomy. Real-world data generated by this observational study address a gap in pediatric perioperative research, where the guidance often relies on extrapolation from adult or higher-risk surgical populations. The results are intended to inform perioperative decision-making and support development of evidence-based fluid management strategies applicable during both routine care and periods of constrained resource availability.

2. Materials and Methods

2.1. Study Design and Data

This retrospective cohort study was conducted at a single tertiary-care pediatric hospital. Pediatric patients undergoing adenotonsillectomy between 1 October 2024 and 31 January 2025 were included. During this period, institutional intraoperative intravenous (IV) fluid administration practices were modified due to a national IV fluid shortage following Hurricane Helene, resulting in a temporary restrictive fluid protocol defined within the institution as <10 mL/kg to sustain the availability of IV fluids, given the institution’s patient volume.

2.2. Study Participants

Eligible patients included children younger than 12 years who underwent adenotonsillectomy during the study period. Patients were identified using operative scheduling records and electronic medical records. Exclusion criteria included incomplete anesthesia documentation, missing postoperative pain assessments, or concomitant major surgical procedures performed during the same anesthetic encounter.

2.3. Measures/Collection

Subjects were stratified based on the total intraoperative IV fluid volume normalized to body weight. This was based on newly implemented institutional practices discussed previously. The cut-off was set to 10 mL/kg, with those under this threshold part of the restricted group and those at or above in the non-restricted group.

Data were extracted via retrospective chart review and included patient demographics, American Society of Anesthesiologists (ASA) physical status classification, intraoperative IV fluid volume, operative duration, presence of nausea and vomiting, postoperative pain scores, and documented postoperative complications. Postoperative pain was assessed using the Face, Legs, Activity, Cry, Consolability (FLACC) scale in the postoperative period upon arrival to the post-operative anesthesia care unit (PACU). FLACC pain scores were recorded by staff only during the postoperative PACU admission. Documented postoperative complications were identified within 7 days of surgery, as early postoperative complications related to perioperative fluid management and oral intake typically occur during the first postoperative week. This was done through review of postoperative documentation, telephone follow-up notes, and any emergency department visits or readmissions documented in the electronic medical record. All data were deidentified before analysis.

2.4. Statistical Analysis

Continuous variables were summarized as medians with interquartile ranges (IQR). The normality of the data distributions was assessed using the Shapiro–Wilk test. However, normality was unable to be achieved, and the nonparametric Mann–Whitney U test was employed to assess postoperative FLACC pain scores, age, sex, weight, operative duration, and intraoperative fluid volumes. Categorical variables, including sex, ASA physical status classification, presence of postoperative nausea and vomiting, and occurrence of postoperative complications, were compared using Fisher’s exact test due to low event counts. All statistical tests were two-sided, and a p-value < 0.05 was considered statistically significant. Statistical analyses were performed using standard statistical software.

2.5. Ethical Considerations

The study was conducted in accordance with the Declaration of Helsinki. Ethical review and approval were given by the Institutional Review Board of the tertiary hospital. IRB: STUDY00000098.

3. Results

3.1. Patient Characteristics

A total of 133 patients were included: 63 in the fluid-restricted group (<10 mL/kg) and 70 in the non-restricted group (≥10 mL/kg) (

Table 1. Baseline demographic and perioperative characteristics of pediatric patients undergoing adenotonsillectomy.

Variable	Fluid-Restricted (<10 mL/kg) (n = 63)	Non-Restricted (≥10 mL/kg) (n = 70)	p-Value
Age, years (median [IQR])	9.0 [7.0–10.0]	7.0 [5.0–8.0]	<0.001
Sex, male n (%)	23 (36.5%)	45 (64.3%)	0.024
Weight, kg (median [IQR])	37.6 [27.8–55.3]	24.7 [19.4–34.0]	<0.001
ASA class I–II, n (%)	51 (81.0%)	61 (87.1%)	0.351
Time in OR, minutes (median [IQR])	51.0 [44.0–63.0]	53.0 [43.2–63.5]	0.869
Total intraoperative fluid, mL (median [IQR])	250 [150–300]	450 [300–500]	<0.001
Weight-adjusted total intraoperative fluid, mL/kg (median [IQR])	6.9 [4.4–8.4]	15.5 [12.9–18.5]	<0.001

Baseline demographic and perioperative characteristics of pediatric patients undergoing adenotonsillectomy, stratified by intraoperative intravenous (IV) fluid strategy. Continuous variables are reported as a median [IQR] and were compared using nonparametric Mann–Whitney U test. Categorical variables are reported as number (%) and were compared using Fisher’s exact test (2 × 2 comparisons). IV = intravenous; ASA = American Society of Anesthesiologists physical status classification.

). The fluid-restricted group was older on average (9.0 [7.0–10.0] vs. 7.0 [5.0–8.0] years, p < 0.001) and had a higher median weight (37.6 [27.8–55.3] vs. 24.7 [19.4–34.0] kg, p < 0.001). Sex distribution differed between groups, with a lower proportion of males in the fluid-restricted cohort (36.5% vs. 64.3%, p = 0.024). The ASA class distribution was similar (ASA I-II: 81.0% vs. 87.1%, p = 0.351), and the time in the operating room did not differ significantly between groups (51.0 [44.0–63.0] vs. 53.0 [43.2–63.5] minutes, p = 0.869). As expected, the weight-adjusted intraoperative fluid administration differed substantially between cohorts (mL/kg: 6.9 [4.4–8.4] vs. 15.5 [12.9–18.5], p < 0.001).

3.2. Postoperative Pain and Complications

Postoperative pain scores (FLACC after) were similar between the fluid-restricted and non-restricted groups, with both cohorts exhibiting an identical median FLACC score of 5.0 (IQR [2.5–6.0] vs. [0.0–7.0], respectively; p = 0.415 (

Table 2. Postoperative pain and complication outcomes by intraoperative fluid strategy.

Outcome	Fluid-Restricted (<10 mL/kg) (n = 63)	Non-Restricted (≥10 mL/kg) (n = 70)	p-Value
FLACC after (median [IQR])	5.0 [2.5–6.0]	5.0 [0.0–7.0]	0.415
Any postoperative complication, n (%)	5 (7.9%)	6 (8.6%)	1.000
Any intraoperative opioid use, n (%)	52 (82.5%)	62 (88.6%)	0.335

Postoperative pain, short-term complication outcomes, and intraoperative opioid use stratified by intraoperative IV fluid strategy. Postoperative pain was assessed using the Face, Legs, Activity, Cry, Consolability (FLACC) scale in the immediate postoperative period. FLACC scores were compared using nonparametric Mann–Whitney U test. Categorical variables are reported as number (%) and were compared using Fisher’s exact test.

)). The overall incidence of documented postoperative complications was low and comparable between cohorts (7.9% vs. 8.6%, p = 1.000). Intraoperative opioid use was frequent in both the restricted (82.5%) and non-restricted (88.6%) groups, with no statistically significant difference in administration rates (p = 0.335).

The single patient in the non-restricted group who developed dehydration with metabolic acidosis was a 9-year-old female patient who presented 7 days after adenotonsillectomy with decreased oral intake due to significant throat pain. Symptoms included reduced eating and drinking since surgery, drooling, nausea, poor sleep, and inability to tolerate solid foods. Initial laboratory studies demonstrated hypoglycemia (glucose 43 mg/dL) and severe anion gap metabolic acidosis with bicarbonate 7 mmol/L and anion gap 24. The clinical picture was consistent with dehydration and starvation-related metabolic acidosis due to inadequate oral intake following surgery. The patient received 325 mL of intravenous normal saline at 14.5 mL/kg during the 40-minute procedure. Complication types were heterogeneous and are summarized descriptively in

Table 3. Descriptive breakdown of postoperative complications.

Complication Type	Fluid-Restricted (n = 63)	Non-Restricted (n = 70)
Febrile episode	1	0
Nausea/Vomiting	3	0
Oropharyngeal pain/PO intolerance	1	1
Dysphonia	0	1
Epistaxis	1	1
Oxygen desaturation/hypoxia	0	1
Hematemesis/bleeding	0	3
Dehydration with metabolic acidosis	0	1

Descriptive breakdown of postoperative complications observed in pediatric patients undergoing adenotonsillectomy, stratified by intraoperative IV fluid strategy. Due to low event counts and heterogeneity of complications, no formal statistical comparisons were performed.

. In the fluid-restricted group, events included febrile episodes, vomiting, oropharyngeal pain, PO intolerance, and other nonspecific presentations. In the non-restricted group, complications included hematemesis, oxygen desaturation, hypoxia, hemorrhage, and dehydration with metabolic acidosis.

4. Discussion

This observational study, precipitated by a hurricane-related IV fluid shortage, provides novel insight into the clinical effects of intraoperative fluid restriction in pediatric adenotonsillectomy. Our findings demonstrate that restricting intraoperative IV fluid administration to less than 10 mL/kg was not associated with significant differences in immediate postoperative pain scores or overall short-term complication rates compared with a non-fluid-restricted approach. These findings suggest that, in the context of short-duration low-risk pediatric procedures, moderate intraoperative fluid restriction may be tolerated without measurable differences in immediate postoperative pain or short-term complication rates.

Perioperative fluid management has increasingly shifted toward restrictive or goal-directed strategies, driven by evidence linking liberal fluid administration to adverse outcomes, including delayed recovery, increased postoperative complications, and longer hospital stays [8,11]. While much of this literature focuses on major abdominal or neonatal surgery, pediatric anesthesia guidelines similarly emphasize the avoidance of fluid overload and recommend balanced isotonic crystalloids with modest glucose supplementation to maintain metabolic homeostasis [2,3,4]. Our findings align with growing evidence that excessive fluid administration offers limited benefit in short-duration pediatric surgeries such as adenotonsillectomies and may not meaningfully influence pain or recovery trajectories.

Postoperative pain following adenotonsillectomy is multifactorial, influenced by the surgical technique, anesthetic approach, airway manipulation, and analgesic regimens. While dehydration has been postulated to exacerbate postoperative pain or nausea, particularly in pediatric patients with prolonged preoperative fasting, our data do not support a clinically significant association between moderate intraoperative fluid restriction and immediate postoperative pain as measured by FLACC scores [14].

Although the overall complication rates were similar between the fluid-restricted and non-fluid-restricted groups, the qualitative pattern of complications differed. The fluid-restricted group experienced events such as febrile episodes, vomiting, and transient intolerance of oral intake, whereas the non-restricted group exhibited complications including hematemesis, hemorrhage, oxygen desaturation, and dehydration with metabolic acidosis. These heterogeneous findings underscore that postoperative complications in adenotonsillectomy are multifactorial and may not be directly attributable to intraoperative fluid volume alone. Importantly, the absence of a consistent or directional signal linking higher or lower fluid volumes to specific adverse outcomes suggests that moderate restriction within clinically accepted limits does not confer excess short-term risk.

From a systems perspective, this study adds to the limited pediatric literature evaluating perioperative care adaptations implemented under resource-limited conditions, including disaster-related disruptions to standard clinical practice. Disaster-induced shortages create unavoidable deviations from standard practice, yet they also offer opportunities to reassess entrenched clinical assumptions. Our findings suggest that for pediatric adenotonsillectomy, the temporary fluid restriction imposed by external constraints did not compromise patient safety or pain control. This has implications not only for disaster preparedness but also for broader discussions around value-based care and avoidance of unnecessary interventions.

Limitations and Future Directions

Several limitations of this study are inherent to its design, including a short implementation window, limited sample size, and the absence of standardized perioperative protocols due to the abrupt disaster-driven nature of the practice change.

Baseline differences were observed between groups, including age, weight, and sex distribution, which may reflect clinical decision-making patterns influencing intraoperative fluid administration. Older and heavier patients may have been more likely to receive lower weight-normalized fluid volumes, introducing potential confounding that could not be fully adjusted for given the study’s retrospective design and limited sample size. Although the operative duration and ASA classification were similar between groups, residual confounding related to patient characteristics cannot be excluded and may have influenced postoperative outcomes. Another limitation included the lack of stratification based on intravenous fluid type due to the lack of available data in the pre-existing collected data.

Additionally, this study’s retrospective design inherently limits causal inference and is subject to unmeasured confounding. Postoperative pain assessment relied on FLACC scores documented in routine clinical care, which may be influenced by observer variability, timing of assessment, and concurrent interventions. Use of FLACC pain scores limits the ability to continuously evaluate pain because it represents a snapshot of pain rather than a longitudinal pain trajectory. In addition, biases due to environmental circumstances in the PACU setting may be present, which include but are not limited to the effects of emergence agitation in the PACU and frequent administration of analgesics in the PACU for pain control. Preoperative NPO duration could not be extrapolated from the data set, which limits the measurement of the baseline hydration status in this analysis. Postoperative complications were identified through chart review and may underestimate minor or delayed events that did not prompt additional clinical encounters. Furthermore, the study was not powered to detect differences in rare but clinically significant complications. The results of this study reflect the experience of a single institution during a time-limited disaster-driven change in practice, which may limit the generalizability to other settings or elective fluid-restriction protocols implemented under different circumstances.

Future prospective studies are needed to more precisely evaluate the relationship between intraoperative fluid strategies and pediatric postoperative outcomes, including standardized pain assessments over time and complication tracking. This study did not analyze intraoperative fluid administration normalized to body weight and time (mL/kg/h). Because most procedures were less than one hour in duration, normalizing fluid administration to mL/kg/h would require extrapolation from short operative times. In this setting, dividing by small fractions of an hour may inflate calculated rates and magnify minor differences in the total administered volume, potentially overstating clinically insignificant variation in fluid management. Further studies may benefit from investigating procedures with longer operative times to address this. Additionally, future prospective studies may benefit from standardized assessment of the NPO timing, permitted oral intake, and post-operative pain management in the PACU to better evaluate its potential role as a confounding variable in perioperative fluid management and postoperative outcomes.

5. Conclusions

This study examines the clinical impact of intraoperative IV fluid restriction in pediatric adenotonsillectomy following a natural disaster that temporarily enacted a change in perioperative fluid management. Reduced fluid administration was not associated with worse postoperative pain outcomes or increased short-term post-operative complications. These findings may suggest that pediatric patients may tolerate lower intraoperative fluid volumes without significantly measurable adverse outcomes in the immediate postoperative period.

This also contributes to the limited pediatric literature evaluating “under-fluid” intraoperative strategies in common short-duration otolaryngologic procedures, particularly under real-world resource-constrained conditions. Further prospective and broader scoped studies with standardized perioperative documentation (including NPO duration, analgesic regimens, and longitudinal pain assessments) are needed to clarify patient-specific factors that may modify risk and to better help with current understanding of optimal perioperative fluid strategies in pediatric surgeries.