The organ-protective effects of nitric oxide in adult patients undergoing cardiac surgery with cardiopulmonary bypass: a systematic review and meta-analysis

Abstract

Background

Postoperative organ dysfunction remains a major challenge in adult cardiac surgery with cardiopulmonary bypass (CPB), frequently involving the kidneys, heart, and lungs. These complications are primarily driven by hemolysis, ischemia-reperfusion injury, and systemic inflammation triggered by CPB. Nitric oxide (NO), known for its vasodilatory, anti-inflammatory, and antioxidant properties, has been proposed as a perioperative strategy to protect vital organs. However, evidence regarding its efficacy remains inconclusive.

Methods

We followed PRISMA guidelines and systematically searched PubMed, Embase, Cochrane Library, and Web of Science for randomized controlled trials (RCTs) published up to March 1, 2025. Subgroup analyses were conducted based on NO dosage and timing of administration. To explore potential effect modifiers and assess subgroup interaction, we performed meta-regression analyses. The GRADE approach was used to assess the certainty of evidence. Sensitivity analyses and publication bias assessments (funnel plots and trim-and-fill method) were also conducted to evaluate the robustness of the findings.

Results

Ten RCTs involving 838 patients were included. NO administration was associated with a reduced incidence of acute kidney injury (AKI) (RR: 0.78; 95% CI: 0.64–0.94; p = 0.010), and the effect remained after trim-and-fill adjustment. mechanical ventilation (MV) duration was slightly shortened (SMD: − 0.17; 95% CI: − 0.31 to − 0.02; p = 0.025), particularly with postoperative administration (SMD: − 0.39; 95% CI: − 0.67 to − 0.12; p = 0.005). NO also reduced cardiac troponin I (cTnI) levels. No significant effects were observed for low cardiac output syndrome (LCOS), mortality, intensive care unit (ICU) length of stay (LOS), or hospital LOS.

Conclusion

Inhaled NO may offer organ-specific benefits in adults undergoing cardiac surgery with CPB, such as reduced AKI incidence and lower cTnI levels. However, these effects did not consistently translate into improved clinical outcomes. The observed reduction in MV duration was not significant after adjusting for publication bias, suggesting a possible overestimation. Current evidence is limited by small sample sizes and small-study effects. Further large, high-quality trials in high-risk populations are needed to confirm these findings.

PROSPERO registration

This review was prospectively registered in PROSPERO (ID: CRD42025649095).

Supplementary Information

The online version contains supplementary material available at 10.1186/s12871-025-03207-7.

Background

Postoperative organ dysfunction is a frequent complication of cardiac surgery with cardiopulmonary bypass (CPB), most commonly affecting the kidneys, heart, and lungs. These complications are largely driven by CPB-induced hemolysis, ischemia-reperfusion injury, and systemic inflammation, contributing to significant perioperative morbidity and healthcare burden [1]. To address these problems, inhaled nitric oxide (NO) has been proposed as a perioperative intervention to protect vital organs during CPB, although its clinical effectiveness remains inconclusive.

In adults undergoing cardiac surgery, the incidence of acute kidney injury (AKI) reaches up to 40% [2–4], low cardiac output syndrome (LCOS) occurs in 5–15% [5–7], and pulmonary injury in approximately 9% [8]. Among pharmacologic strategies explored to mitigate such complications, NO has gained attention due to its vasodilatory, anti-inflammatory, and antioxidant properties. It improves endothelial function via the soluble guanylate cyclase (sGC)/cyclic guanosine monophosphate (cGMP) pathway, enhancing microcirculatory perfusion and limiting oxidative damage [9, 10]. Clinically, Inhaled NO is utilized for pulmonary hypertension (PH) and acute respiratory distress syndrome (ARDS) by reducing right ventricular afterload and enhances oxygenation [11–13]. In cardiac surgery, the administration of inhaled NO mitigates the depletion of endogenous NO caused by cell-free hemoglobin (CFHb), preserving renal and myocardial function [14–16].

Despite its potential benefits, the use of NO in cardiac surgery remains controversial. Recent randomized controlled trials (RCTs) have reported inconsistent findings regarding the renal protective effects of NO [16–18]. Meanwhile, meta-analyses have shown no significant benefit in preventing pulmonary complications such as PH and ARDS [19, 20]. These conflicting findings highlight the need for further investigation into its clinical utility and optimal application strategy. Despite its proposed benefits, a comprehensive synthesis of nitric oxide’s organ-protective effects across the renal, cardiac, and pulmonary systems is still lacking. Furthermore, critical clinical questions such as the optimal timing and dosing of NO administration remain unanswered [19–21]. Therefore, an updated systematic review integrating evidence across multiple endpoints is warranted to clarify its therapeutic potential in this setting.

This study aims to systematically evaluate the organ-protective and prognostic effects of NO in adult patients undergoing cardiac surgery with CPB. We specifically assess its impact on renal, cardiac, and pulmonary outcomes, measured by AKI, LCOS, the level of cardiac troponin I (cTnI) and mechanical ventilation (MV) duration, as well as clinical prognosis, including intensive care unit (ICU) length of stay (LOS), hospital LOS, and mortality. The analysis further investigates the potential influence of timing and dosage on NO’s effectiveness.

Methods

This study followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) [22]. A completed PRISMA checklist with page numbers for each item is provided in Supplementary Material 1. The protocol of this study was registered at PROSPERO (ID: CRD42025649095).

Inclusion and exclusion criteria

The inclusion criteria for the studies were as follows. (1) Participants: adult patients aged 18 year or older undergoing cardiac surgery with CPB. (2) Intervention: NO administered intraoperatively or postoperatively, regardless of the type of administration method or dose. (3) Comparison: comparator groups receiving placebo, no treatment, or standard care. (4) Outcomes: Studies were required to report at least one of the following outcomes including AKI, MV duration, LCOS, and the level of cTnI. (5) Study design: RCTs.

The exclusion criteria for the studies were as follows: (1) case reports, case series, nonrandomized studies, observational studies, or review articles; (2) studies involving pediatric patients; (3) studies comparing other vasodilators without a control group; (4) studies not reporting the outcomes of interest or providing insufficient data for the calculation of effect sizes; and (5) duplicate publications or studies using the same patient population. In such cases, the most comprehensive or most recent publication was included.

Search strategy

We searched PubMed, Embase, the Cochrane Library, and Web of Science for all RCTs comparing inhaled NO with placebo, standard care, or no treatment in adult patients undergoing CPB, published up to March 1, 2025. The search was limited to English-language publications. Detailed search strategies are provided in Supplementary Material 2. Two independent reviewers screened titles and abstracts, and assessed full texts based on predefined inclusion and exclusion criteria. Discrepancies were resolved by consensus or by consulting a third reviewer.

Data extraction

Two independent reviewers extracted data on study characteristics (first author, publication year, study design, sample size), patient demographics (age, sex, baseline comorbidities), and surgical details (type of surgery, duration of surgery, duration of CPB). Information on the NO intervention was also collected, including method of administration, dosage, and timing of delivery.

Outcomes were categorized into four domains: (1) Renal outcomes, including the incidence of AKI; (2) Pulmonary outcomes, assessed by the duration of MV; (3) Cardiac outcomes, including cTnI levels and the incidence of LCOS; (4) Prognostic outcomes, including mortality, LOS in the ICU, and total hospital LOS.

Statistical analysis

All statistical analyses were performed using R (version 4.4.2) with the “meta”, “metabin”, “metacont”, “metaif” packages. We conducted a meta-analysis for outcomes reported by at least two studies. For continuous outcomes (MV duration, ICU LOS, hospital LOS, and cTnI levels), the standardized mean differences (SMDs) with 95% confidence intervals (CIs) were calculated. For binary outcomes (mortality, AKI, LCOS), odds ratios (ORs) and risk ratios (RRs) with 95% CIs were computed. Heterogeneity across studies was assessed using Cochran’s Q test and the I² statistic. I² value greater than 50% indicated substantial heterogeneity. A random-effects model was applied when significant heterogeneity was present; otherwise, a common-effects model was used. The Hartung-Knapp adjustment was applied to account for uncertainty in the estimation of between-study variance. Subgroup analyses were conducted to assess the effects of NO dosage and administration timing on the outcomes. NO dosage was categorized as < 40 ppm, 40–80 ppm (including 40 and 80), and > 80 ppm. Administration timing was classified as intraoperative only, postoperative only, or continuous from intra- to postoperative periods. Statistical interaction between subgroups was tested using χ²-based Q tests for subgroup differences. Meta-regression analyses were conducted to explore potential effect modifiers and assess subgroup interaction. Sensitivity analyses were performed to assess the robustness of the results by sequentially excluding each study and recalculating the pooled effect size. Publication bias was assessed visually using funnel plots and statistically using the trim-and-fill method, which was also applied to estimate the adjusted effect size when asymmetry was present. The significance level was set at p < 0.05.

Quality assessment

All included studies were RCTs. The risk of bias was assessed using the Cochrane Risk of Bias 1.0 (ROB 1.0) tool. Two reviewers independently conducted the assessments, and any discrepancies were resolved through discussion or by consulting a third reviewer. The results of the risk-of-bias evaluations were visualized using the robvis package in R, providing both traffic-light plots and summary figures.

Certainty of evidence assessment

Each outcome was assessed using the Grading of Recommendations, Assessment, Development and Evaluation (GRADE) approach [23]. A summary of findings table was generated using GRADEpro GDT (McMaster University, Ontario, Canada; available at https://gradepro.org).

Results

Search results and trial characteristics

Our systematic search identified 4,246 records from multiple databases, of which 31 studies underwent full-text eligibility assessment (Fig. 1). After screening, 31 studies underwent full-text review, and 10 RCTs involving 838 patients were included in the final meta-analysis (Tables 1 and 2) [14–17, 24–29]. These studies were published between 1998 and 2025, with individual sample sizes ranging from 29 to 244 participants. The most common NO dosage was 40 ppm, although higher dosage has been employed in recent trials. Regarding the timing of NO administration, four studies administered NO postoperatively, another four intraoperatively, and two studies applied it continuously from the intraoperative period through to the postoperative period.

Fig. 1.

Flow chart for the study selection

Table 1.

All the studies included in our meta-analysis

Study	Year	Sample		Surgery type	dosage (ppm)	administration route
		NO	placebo
Kamenshchikov et al., 2025 [25]	2025	37	37	CABG, valve surgery	200	respiratory circuit
Azem et al., 2024 [17]	2024	51	47	CABG, valve surgery	40	CPB circuit
Kamenshchikov et al., 2022 [16]	2022	48	48	CABG, valve surgery	40	CPB circuit
Pichugin et al., 2020 [26]	2020	30	30	valve surgery	20	respiratory circuit
Kamenshchikov et al., 2019 [15]	2019	30	30	CABG	40	CPB circuit
Lei et al., 2018 [14]	2018	117	127	valve surgery	80	CPB circuit and respiratory circuit
Potapov et al., 2011 [29]	2011	69	68	LVAD implantation	40	respiratory circuit
Fernandes et al., 2011 [27]	2011	20	20	valve surgery	20	respiratory circuit
Gianetti et al., 2004 [28]	2004	14	15	cardiac surgery with CPB	40	respiratory circuit
Prendergast et al., 1998 [24]	1998	14	15	CABG	10–40	respiratory circuit

NO nitric oxide, CABG coronary artery bypass grafting, LVAD left ventricular assist device, CPB cardiopulmonary bypass

Table 2.

Subgroup analysis by timing and dosage of inhaled NO across different outcomes

Outcome	Subgroup	No. of studies	Effect (95% CI)	Heterogeneity (I²), %	P value (test for effect)
AKI	Dosage (ppm)
	< 40	1	0.36 [0.02; 8.07]	-	-
	40–80	5	0.79 [0.65; 0.96]	12	0.017
	> 80	1	0.70 [0.30; 1.64]	-	-
	Timing
	Intraoperative	3	0.69 [0.20, 2.39]	42.2	0.10
	Continuous	1	0.78 [0.62, 0.97]	-	-
	Postoperative	3	0.89 [0.33, 2.43]	0	0.71
	Overall	7	0.78 [0.65; 0.94]	0	0.010
MV duration	Dosage (ppm)
	< 40	3	−0.38 [−0.79; 0.02]	0	0.06
	40–80	5	−0.13 [−0.29; 0.02]	22.3	0.09
	Timing
	Intraoperative	3	−0.14 [−0.89; 0.61]	43.8	0.35
	Continuous	2	−0.04 [−0.23; 0.14]	0	0.72
	Postoperative	3	−0.39 [−0.67; −0.12]	0	0.005
	Overall	8	−0.17 [−0.31; −0.02]	13.9	0.025
ICU LOS	Dosage (ppm)
	< 40	3	−0.30 [−0.70; 0.10]	0	0.14
	40–80	5	−0.16 [−0.57; 0.24]	62.9	0.32
	Timing
	Intraoperative	3	−0.32 [−1.30; 0.67]	71.6	0.30
	Continuous	2	0.00 [−0.24, 0.24]	0	1.00
	Postoperative	3	−0.13 [−0.40, 0.15]	34.2	0.36
	Overall	8	−0.14 [−0.29; 0.00]	46.8	0.06
Hospital LOS	Dosage (ppm)
	< 40	2	−0.44 [−0.91; 0.04]	0	0.07
	40–80	5	−0.09 [−0.24; 0.07]	0	0.28
	> 80	1	−0.88 [−1.36; −0.40]	-	-
	Timing
	Intraoperative	3	−0.21 [−0.46; 0.03]	13	0.09
	Continuous	1	0.00 [−0.25; 0.25]	-	-
	Postoperative	4	−0.41 [−1.04; 0.22]	66.9	0.13
	Overall	8	−0.25 [−0.52; 0.01]	53.1	0.06

NO nitric oxide, AKI acute kidney injury, MV mechanical ventilation, ICU intensive care unit, LOS length of stay, CI confidence interval

Risk of bias assessment

The risk of bias for the included studies was assessed using the Cochrane Risk of Bias Tool. Approximately half of the studies were judged to be at low risk across most domains. However, four studies were rated as having high or unclear risk of bias in one or more domains. The most commonly observed concern was performance bias due to the lack of blinding of participants and personnel. Additional issues included unclear reporting of allocation concealment and random sequence generation. Detailed results of the risk of bias assessment are presented in Fig. 2.

Fig. 2.

Risk of bias for each study

Outcomes

AKI

Seven studies (n = 738) reported the incidence of AKI. Pooled analysis suggested that NO administration may be associated with a reduced risk of AKI compared to control (RR: 0.78; 95% CI: 0.64–0.94; p = 0.010), with no observed heterogeneity (I² = 0%) (Fig. 3).

Fig. 3.

Forest plot of NO on the incidence of AKI. Note: NO, nitric oxide; AKI, acute kidney injury; CPB, cardiopulmonary bypass; CI, confidence interval; RR, risk ratio

MV duration

A total of eight studies involving 733 patients were included in the analysis. Compared with the control group, NO administration was initially associated with a modest reduction in MV duration (SMD: − 0.17; 95% CI: − 0.31 to − 0.02; p = 0.025), with low heterogeneity (I² = 13.9%) (Fig. 4).

Fig. 4.

Forest plot of NO on the duration of MV. Note: NO, nitric oxide; MV, mechanical ventilation, CPB, cardiopulmonary bypass; CI, confidence interval

LCOS and cTnI levels

To explore the potential cardioprotective effects of NO, we assessed both the incidence of LCOS and levels of cTnI, a biomarker of myocardial injury. NO administration did not significantly reduce the incidence of LCOS (RR: 1.48; 95% CI: 0.60–3.67; p = 0.39). However, a significant reduction in cTnI levels was observed in the NO group (SMD: − 0.63; 95% CI: − 1.00 to − 0.26; p < 0.001), with no heterogeneity (I² = 0.0%) (Fig. 5). Given that this result was based on only two studies and that publication bias could not be reliably assessed, the finding should be interpreted with caution.

Fig. 5.

Forest plot of NO on the cardiac protection. Note: A Forest plot showing the overall effect of NO on the level of cTnI. B Forest plot showing the overall effect of NO on the incidence of LCOS. NO, nitric oxide, cTnI, cardiac troponin I, LCOS, low cardiac outcome surgery, CPB, cardiopulmonary bypass; CI, confidence interval; SMD, standardized mean difference; RR, risk ratio

Mortality, ICU LOS and hospital LOS

The meta-analysis found no statistically significant effect of NO on mortality (OR: 0.74; 95% CI: 0.36–1.50; p = 0.42), with no observed heterogeneity (I² = 0.0%). Similarly, NO did not significantly reduce ICU LOS (SMD: − 0.14; 95% CI: − 0.29 to 0.00; p = 0.058; I² = 46.8%) or hospital stay duration (SMD: − 0.25; 95% CI: − 0.52 to 0.01; p = 0.059; I² = 53.0%) (Supplementary Material 3).

Subgroup and sensitivity analyses

Subgroup analyses were performed for AKI, MV duration, ICU stay, and hospital stay based on NO dosage and timing of administration. A potential benefit was observed in the 40–80 ppm dosage subgroup for AKI (RR: 0.79; 95% CI: 0.65–0.96; p = 0.017) and in the postoperative administration subgroup for MV duration (SMD: − 0.39; 95% CI: − 0.67 to − 0.12; p = 0.005) (Supplementary Material 4). However, these findings should be interpreted with caution due to small sample sizes in some subgroups. Other subgroup comparisons did not yield statistically significant differences. Sensitivity analyses showed that the overall findings were robust to the exclusion of individual studies (Supplementary Material 5).

Publication bias assessment

Funnel plot inspection indicated slight asymmetry for several outcomes, suggesting the possibility of publication bias (Supplementary Material 6). The trim-and-fill method imputed missing studies for AKI (n = 2), MV duration (n = 2), ICU stay (n = 2), and hospital stay (n = 3). After adjustment, the associations for AKI remained statistically significant, though with a reduced effect size. In contrast, the effects for MV duration, ICU stay, and hospital stay were no statistically significant post-adjustment. These findings indicate that some observed associations may have been influenced by small-study effects or selective publication, particularly for outcomes with marginal significance.

Certainty of evidence

According to the GRADE assessment, high-certainty evidence supported the reduction in AKI. The certainty of evidence was low for MV duration, cTnI, ICU LOS, and hospital LOS, mainly due to risk of bias, heterogeneity, and imprecision. Mortality was supported by moderate-certainty evidence, while the certainty for LCOS was very low due to serious imprecision and limited data (Supplementary Material 7).

Meta-regression analysis

Meta-regression analysis suggested that postoperative NO administration might be associated with a shorter duration of MV. For other outcomes, including AKI, ICU stay, and hospital stay, no statistically significant associations were observed with timing, NO dosage, or publication year. These findings indicate a potential benefit of postoperative NO use on ventilation duration, although the influence on other outcomes remains uncertain (Supplementary Material 8).

Discussion

Our meta-analysis suggests that inhaled NO may confer organ-specific protective effects in adult patients undergoing cardiac surgery, particularly by reducing the incidence of AKI and lowering postoperative cTnI levels. A potential reduction in MV duration was also observed; however, this association lost statistical significance after adjusting for publication bias, indicating a possible influence of small-study effects. Importantly, inhaled NO did not show significant benefits in broader clinical outcomes such as LCOS, ICU/hospital LOS, or mortality.

Our study builds upon prior meta-analyses by incorporating newly published randomized trials, performing a dose-based subgroup analysis, and broadening the scope of investigation to assess the multi-organ protective effects of inhaled nitric oxide rather than focusing solely on renal outcomes [30]. In addition, we conducted rigorous assessments of publication bias and applied the GRADE framework to evaluate the certainty of evidence. Together, these enhancements provide a more comprehensive and up-to-date synthesis of the systemic benefits of inhaled NO across different timepoints and organ systems.

Renal protection

Our findings suggest that NO may reduce the incidence of AKI in adults undergoing CPB by mitigating hemolysis-induced oxidative stress and endothelial dysfunction. CPB-related hemolysis releases CFHb, which rapidly scavenges endogenous NO, leading to vasoconstriction, oxidative injury, and renal hypoperfusion [31–35]. Exogenous NO may counteract these effects by maintaining microvascular tone, reducing inflammation, and protecting mitochondria against reperfusion-induced damage [36–39]. At the cellular level, NO is known to preserve mitochondrial function by inhibiting the opening of the mitochondrial permeability transition pore and activating ATP-sensitive potassium channels, thereby limiting tubular apoptosis.

In parallel, recent evidence has highlighted Amino acid supplementation for cardiac surgery-associated AKI prevention. Amino acid supplementation has emerged as the only intervention supported by a large RCT (the PROTECTION trial), which showed a significant reduction in AKI incidence and is now endorsed by the EACTS/EACTAIC/EBCP 2024 guideline [8, 39].Importantly, the renal protective effects of amino acids are thought to be partially mediated through stimulation of endogenous NO synthesis, especially L-arginine, the substrate for endothelial NO synthase [40]. This mechanistic overlap suggests a potential synergistic interaction between amino acid therapy and exogenous NO administration. While inhaled NO may exert a mild protective effect against AKI, future studies should explore this combination approach, especially in high-risk surgical cohorts.

While the pooled data suggest a renoprotective effect of NO, individual RCTs yielded inconsistent findings, likely due to variations in patient populations, AKI definitions, and NO administration protocols. Studies reporting benefit tended to enroll high-risk patients (e.g., prolonged CPB, preexisting endothelial dysfunction), and used intraoperative continuous NO at 40–80 ppm [14, 16]. Notably, while our subgroup analysis showed a reduction in AKI incidence with 40–80 ppm NO, this pattern did not reach statistical significance in meta-regression, likely due to the limited number of included studies (k = 7) and reduced analytical power of regression in small datasets. We assessed publication bias using the trim-and-fill method. While the adjusted effect size remained statistically significant for AKI, this approach has important limitations, particularly when applied to meta-analyses with fewer than 10 studies. In such contexts, the method may not reliably account for small-study effects, and the results should therefore be interpreted with caution.

Pulmonary protection

Our analysis suggested a slight reduction in the duration of MV with inhaled NO after cardiac surgery; however, this effect was small and its clinical relevance remains uncertain, particularly in light of potential publication bias and small-study effects. Mechanistically, NO activates the cGMP–PKG signaling pathway, lowers pulmonary vascular resistance, and selectively enhances perfusion in well-ventilated lung regions [41]. Clinical studies suggest that inhaled NO may be particularly beneficial in patients with baseline hypoxemia or pulmonary hypertension, in whom improved gas exchange can hasten respiratory recovery [17, 26]. In our subgroup analysis, postoperative NO administration showed a trend toward greater reduction in ventilation duration, possibly because it coincides with the peak of systemic inflammation and reperfusion injury, when exogenous NO is most needed to support endothelial function.

Importantly, after adjustment for potential publication bias using the trim-and-fill method, the effect on ventilation time was no longer statistically significant, indicating a possible influence of small-study effects. Second, the magnitude of the observed reduction in ventilation duration was subtle and may not translate into meaningful clinical benefit. Third, MV duration is highly dependent on institutional weaning protocols and postoperative care standards, which introduces variability that is difficult to adjust for.

Cardiac protection

Although based on a limited number of studies, our analysis suggests a potential association between and reduced cTnI levels, indicating subclinical myocardial protection. However, it did not significantly lower the incidence of LCOS, suggesting a disconnect between biomarker improvement and clinical outcomes. This discrepancy may reflect the differential sensitivity of these endpoints: while cTnI is a sensitive indicator of subclinical myocardial injury, LCOS represents overt hemodynamic dysfunction that requires more substantial myocardial recovery. The reduction in cTnI suggests that NO may attenuate ischemia-reperfusion injury, but this effect may be insufficient to prevent clinically apparent cardiac failure.

Furthermore, the limited number of studies (n = 2) and small sample sizes, especially for LCOS, a relatively infrequent event, restrict the ability to detect meaningful differences. Variability in NO dosing strategies and patient baseline risk may have further obscured potential benefits. To clarify whether biochemical improvements can translate into clinical gains, future trials should adopt standardized NO administration protocols and incorporate pre-specified subgroup analyses, particularly among high-risk patients such as those with prolonged CPB duration or impaired ventricular function.

Prognostic outcomes

Despite some positive trends in renal, pulmonary, and cardiac surrogate markers, these did not translate into statistically significant improvements in hard clinical endpoints such as ICU/hospital LOS or mortality. Several explanations are possible. First, the magnitude of observed effects was relatively small and close to the threshold of significance, which may not be sufficient to influence complex, multifactorial outcomes. These outcomes are influenced by numerous perioperative variables, including patient comorbidities, surgical complexity, intra- and postoperative complications, and institutional discharge criteria, that may dilute or mask the impact of a single intervention like NO. Although NO may mitigate specific injuries, such as renal or myocardial damage, these effects may not be large or consistent enough to independently influence downstream outcomes. Second, in most included studies, mortality and prolonged ICU/hospital stay were not the primary outcomes. As a result, these endpoints were often underreported, and the event rates were generally low. The sample sizes were also limited, reducing the statistical power to detect differences in rare events like death. Thus, a true effect of NO cannot be entirely excluded but may be obscured by insufficient data. Third, our analysis identified moderate heterogeneity in ICU and hospital stay outcomes. Subgroup analyses showed that heterogeneity decreased when stratified by NO administration timing or dose, suggesting that variability in NO protocols may have influenced the results. Differences in surgical types and patient comorbidities across studies may also contribute to increased heterogeneity. Additionally, heterogeneity in perioperative management (e.g., fluid strategies, renal protection protocols, ventilation weaning criteria) across studies may have further obscured potential benefits. These inter-study differences make it difficult to attribute hard outcome effects solely to NO.

Limitation

Several limitations should be acknowledged. First, although inhaled NO was associated with a reduced incidence of AKI, the effect size was small, and its clinical significance remains uncertain. Moreover, while the trim-and-fill method suggested that the association with AKI remained statistically significant after adjusting for potential publication bias, this approach has important limitations, particularly when applied to meta-analyses with fewer than 10 studies. Second, pulmonary outcomes were inconsistently reported across trials, leading to the use of MV duration as a surrogate endpoint. However, this measure is subject to variability in local weaning protocols, clinical practices, and may not reliably reflect lung injury or recovery. Notably, the initially observed benefit in ventilation time was no longer significant after correction for potential publication bias, suggesting the effect may have been overestimated. Third, several key outcomes including LCOS, cTnI levels, and mortality were assessed in only a few small trials, limiting the statistical power and generalizability of these findings. For example, the analysis of cTnI was based on just two studies, and conclusions regarding cardioprotection should be interpreted with caution. Lastly, this meta-analysis was based on aggregate data, preventing more granular patient-level analyses and adjustment for important confounders such as CPB duration, baseline organ function, and comorbidities. These limitations highlight these findings should be interpreted with caution and the need for future well-powered, patient-level studies using standardized NO protocols and clinically meaningful, organ-specific outcomes.

Conclusion

Inhaled NO may offer organ-specific benefits in adults undergoing cardiac surgery with CPB, including a reduction in AKI incidence and lower cTnI levels. However, these physiological effects did not consistently lead to improvements in clinical outcomes such as mortality or ICU/hospital LOS. Importantly, the initially observed reduction in MV duration was no longer significant after adjusting for potential publication bias using the trim-and-fill method, indicating that this effect may be overestimated. The current evidence is limited by small sample sizes and possible small-study effects. Further large-scale, high-quality trials targeting high-risk populations are needed to confirm these findings and clarify the clinical value of NO in cardiac surgery.

Abbreviations

CPB

Cardiopulmonary bypass

NO

Nitric oxide

AKI

Acute kidney injury

MV

Mechanical ventilation

CFHb

Cell free hemoglobin

cTnI

cardiac troponin I

LCOS

Low cardiac output syndrome

ICU

Intensive care unit

LOS

Length of stay

sGC

soluble guanylate cyclase

cGMP

cyclic guanosine monophosphate

PH

Pulmonary hypertension

ARDS

Acute respiratory distress syndrome

PRISMA

Preferred reporting items for systematic reviews and meta-analyses

SMD

Standardized mean difference

OR

Odds ratio

RR

Risk ratio

CI

Confidence interval

ROB

Risk of bias

GRADE

Grading of Recommendations, Assessment, Development and Evaluation

Authors’ contributions

J.Z., and H.Z. systematic search; J.W. and T.W. data extraction; J.Z. and H.Z. writing—original draft preparation; L.B., J.W. and B.J. writing—review and editing; Z.L. and Y.T. prepared figures; G.L. prepared tables; S.Y. and H.Z. prepared supplemental materials; J.Z. software; J.Z. and T.W. PROSPERO registration; B.J. conceptualization and funding acquisition. All authors reviewed the manuscript.

Funding

No Funding.

Data availability

The data has been entirely included in the manuscript and supplementary files. Code for the analyses can be obtained from the corresponding author.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.