Introduction
Serum uric acid (SUA) is the end product of cellular purine metabolism, formed through the oxidation of xanthine and hypoxanthine by xanthine oxidoreductase, a reaction that is essential for the removal of nitrogenous waste products [1]. In humans, normal SUA levels are reported to be 2.6–5.7 mg/dl (155–339 µmol/L) for premenopausal women and 3.5–7.0 mg/dl (208–416 µmol/L) for men and postmenopausal women [1–3]. SUA levels have been extensively studied due to their association with various diseases, including gout, hypertension (HTN), diabetes, and cardiovascular, renal, and metabolic disorders [4–7]. In addition, preoperative SUA levels have been reported to be associated with postoperative mortality and complications across various surgical types and settings [8–11]. However, despite decades of research on the pathophysiological role of uric acid in different disease processes, the influence of perioperative SUA on surgical prognosis remains not fully understood.
Recent studies have shifted the focus from a simplistic linear relationship between preoperative SUA levels and postoperative outcomes to a more nuanced understanding that suggests a J- or U-shaped association with all-cause and cardiovascular mortality [12–15], meaning that both very low and very high SUA levels may elevate the risk of adverse outcomes. Furthermore, it has been reported that SUA levels differ by sex, and the risk thresholds that influence outcomes may also vary by sex [16,17]. These findings imply that populations should be differentiated and stratified according to sex and patient risk when evaluating the impact of SUA on surgical outcomes. However, previous studies have primarily focused on cardiac surgery or specific clinical settings, and research on the impact of preoperative SUA levels on postoperative outcomes in non-cardiac surgery remains limited. Specifically, the effects of SUA on short- and long-term outcomes, including mortality and complications, have not been thoroughly clarified across the general population, different sexes, or high-risk surgical groups.
We hypothesized a priori that the relationship between SUA levels and postoperative outcomes would differ by sex due to known physiological differences in uric acid metabolism. Therefore, this study aimed to investigate the association between sex-specific preoperative SUA levels and postoperative outcomes, including both short- and long-term mortality and morbidity, and to identify the lower-risk SUA range for both men and women in a large single-center cohort of non-cardiac surgery patients. We also evaluated the impact of SUA in high-risk populations, particularly among elderly patients, those with a high American Society of Anesthesiologists (ASA) classification, and patients deemed at high surgical risk according to the European Society of Cardiology (ESC) criteria.
Materials and Methods
Study design and participants
This study was a retrospective analysis utilizing electronic medical record data from our center and received approval from its Institutional Review Board (Asan Medical Center, 2024-1090). Due to the retrospective nature of the study design, written informed consent was waived. We analyzed data from adult patients aged 18 years and older who underwent non-cardiac surgery at our center between January 2012 and December 2021. If multiple surgeries were performed during the study period, only the first surgery following admission was included in our analysis. The exclusion criteria were as follows: emergency surgery, pregnancy, cadavers, and patients with missing preoperative SUA values or no follow-up after surgery.
Data collection, perioperative variables, and study outcomes
The electronic medical record system was retrospectively reviewed to gather data on patients, including baseline characteristics, surgical variables, laboratory values, and surgical outcomes. Baseline characteristics included age, sex, height, weight, body mass index, preoperative medications (e.g., calcium channel blockers, angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, beta-blockers, anticoagulants, statins, diuretics, and chemotherapy), ASA physical status classification, and inpatient status. Information on comorbidities such as gout, kidney stones, diabetes mellitus, HTN, cerebrovascular accidents, cardiovascular disease (CVD), renal disease, respiratory disease, hepatitis, and hyperlipidemia were also collected.
Operation-specific variables included ESC surgical risk [18], type of surgery, high-risk surgery (e.g., orthopedic, neurosurgical, thoracic, vascular), intraoperative red blood cell transfusion, operation time, and cancer surgery. Surgical procedures were classified into general, otolaryngologic, urologic, orthopedic, gynecologic, neurosurgical, thoracic, vascular, and other categories. High-risk surgery was predefined as orthopedic, thoracic, vascular, or neurosurgical procedures, consistent with established perioperative risk stratification frameworks (e.g., Revised Cardiac Risk Index, American College of Cardiology/American Heart Association, ESC/European Society of Anesthesiology guidelines, and the University of California, Los Angeles Surgical Risk Score) [18–20]. Laboratory parameters comprised preoperative SUA levels, white blood cell count, hemoglobin, platelet count, international normalized ratio, total bilirubin, albumin, sodium, potassium, chloride, aspartate aminotransferase, alanine aminotransferase, creatinine, estimated glomerular filtration rate (eGFR) using the Modification of Diet in Renal Disease formula, cholesterol, total protein, total calcium, and glucose levels. Preoperative laboratory tests performed within four weeks prior to surgery, including complete blood counts and SUA levels, were included.
The primary outcome was the association between sex-specific preoperative SUA levels and postoperative mortality—both short-term (30-day, 180-day) and long-term (one-year, overall) in males and females. We also sought to identify the SUA range associated with the lowest risk of postoperative mortality in both sexes.
Secondary outcomes included postoperative morbidity, defined as complications occurring within 30-day after surgery. Morbidity outcomes included intensive care unit (ICU) admission and a composite of complications that encompassed CVD—a combined measure of myocardial infarction, heart failure, and fatal dysrhythmias—along with bleeding, stroke, venous thromboembolism, wound dehiscence, surgical site infection, systemic infection, pneumonia, sepsis, and acute kidney injury (AKI) as defined by the KDIGO criteria [21], as well as urinary tract infection. Postoperative morbidity was identified using data from medical records that included imaging findings (e.g., computed tomography and X-rays), postoperative laboratory results, progress notes, and diagnostic codes from the International Classification of Diseases, 9th and 10th Revisions, in addition to in-hospital diagnostic codes from Asan Medical Center. Furthermore, we evaluated the association of sex-specific preoperative SUA levels stratified by age (< 65 vs. ≥ 65 years), ASA class (1, 2 vs. ≥ 3), and ESC surgical risk (1, 2 vs. 3) with mortality at 30-day, 180-day, one year, and overall, as well as with composite complications.
Statistical analysis
Data are presented as mean ± SD, median (Q1, Q3), or number (proportion), as appropriate. Categorical variables were compared using the chi-squared test or Fisher’s exact test, while continuous variables were analyzed using Student’s t-test or the Mann–Whitney U-test. Cox proportional hazards regression and multivariate logistic regression analyses were performed to identify risk factors influencing survival at 30-day, 180-day, one year, and throughout the overall follow-up period, as well as to assess postoperative morbidity. Variables with a P value below 0.1 in the univariate analysis were included in the multivariate analysis. For the risk factor analysis, SUA levels were categorized into one-unit intervals, resulting in seven groups based on the distribution of SUA levels: < 4 to ≥ 9 for males and < 3 to ≥ 8 for females. SUA categories were defined as left-inclusive and right-exclusive intervals (e.g., 4–5 mg/dl includes values ≥ 4.0 and < 5.0 mg/dl), and this rule was applied consistently for all groups. The reference category for SUA levels was defined separately for males and females as the interval with the lowest observed mortality rate, considering 30-day, 180-day, one-year, and overall mortality collectively.
To evaluate the nonlinear relationship between preoperative SUA levels and postoperative mortality (180-day, one-year, and overall mortality), we conducted a multivariable analysis utilizing natural cubic splines within a Cox proportional hazards model. The analysis excluded 30-day mortality to prevent potential overfitting. SUA levels were trimmed at the 1st and 99th percentiles to eliminate outliers, and spline knots were positioned at the 25th, 50th, and 75th percentiles of SUA levels to enhance model flexibility. Nonlinear relationships were assessed using likelihood ratio tests, with P values reported for overall effects (P for overall) and nonlinearity (P for nonlinear). The hazard ratio (HR) was standardized to the minimum observed HR, and 95% CIs were calculated based on the standard error of predictions. The lower-risk SUA range was defined as the interval of SUA levels where the 95% CI for HR included 1, and the reference SUA level was identified as the value closest to an HR value of 1.
Kaplan–Meier curves were employed to visualize survival trends and compare outcomes based on SUA levels in both males and females. Subgroup analyses were conducted to assess the prognostic value of SUA levels across various factors, including age (< 65 vs. ≥ 65 years), ESC surgical risk (low/intermediate vs. high), ASA classification (1/2 vs. 3), and baseline renal function as defined by eGFR (< 60 vs. ≥ 60 ml/min/1.73 m2). All reported P values were two-sided, with P < 0.05 deemed statistically significant. Statistical analyses were performed using SAS version 9.4 (SAS Institute Inc.) and R version 4.3.1 (R Foundation for Statistical Computing).
Results
Baseline characteristics and perioperative variables according to sex
A total of 309 353 patients aged 18 years or older who underwent non-cardiac surgery were identified from the medical records. Of these, 14 086 patients were excluded. After exclusions, 295 267 patients (female, n = 164 756 [55.8%]) were included in the final analysis (Supplementary Fig. 1). The demographic, operative, laboratory variables, and study outcomes for males and females are presented in Tables 1 and 2, respectively, with additional details provided in Supplementary Tables 1 and 2.
Significant differences were observed across all groups of males and females, particularly among the SUA groups, in terms of demographic variables, medication use, comorbidities, operative variables, and laboratory results (all P < 0.001). The reference range for SUA levels was established based on the interval associated with the lowest mortality rate, set at 6 to 7 mg/dl for males and 4 to 5 mg/dl for females.
Surgical outcomes according to sex
The 30-day, 180-day, one-year, and overall mortality rates for males were 0.2%, 1.6%, 3.6%, and 16.0%, respectively (Table 1). The corresponding rates for females were 0.1%, 0.8%, 1.8%, and 8.3%, respectively (Table 2), with all comparisons demonstrating significant differences across SUA groups (P < 0.001 for each time point). The composite complication rate was 9.2% in males and 3.8% in females, with significant differences noted across all SUA groups (P < 0.001) (Tables 1 and 2).
In males, Cox regression analysis showed SUA levels were significantly associated with increased mortality at 30-day (P = 0.004), 180-day (P < 0.001), one year (P < 0.001), and overall (P < 0.001) (Table 3). Compared to the reference group (6–7 mg/dl), SUA < 4 and ≥ 9 mg/dl were linked to higher short- and long-term mortality risks. In the multivariate logistic regression analysis examining risk factors for morbidity, SUA levels were significantly associated with composite complications (P < 0.001), bleeding (P = 0.004), and AKI (P < 0.001). SUA levels of 4–5 mg/dl (OR: 0.88, 95% CI [0.82–0.94], P < 0.001) and 5–6 mg/dl (OR: 0.92, 95% CI [0.87–0.98], P = 0.012) were associated with a reduced risk of composite complications, whereas SUA levels < 4 mg/dl (OR: 1.07, 95% CI [1.00–1.15], P = 0.043), 8–9 mg/dl (OR: 1.15, 95% CI [1.03–1.28], P = 0.015), and ≥ 9 mg/dl (OR: 1.18, 95% CI [1.02–1.36], P = 0.023) were associated with an increased risk of composite complications. Similarly, SUA levels 4–5 mg/dl (OR: 0.83, 95% CI [0.77–0.90], P < 0.001), and 5–6 mg/dl (OR: 0.93, 95% CI [0.86–1.00], P = 0.041) were associated with a reduced risk of AKI. In contrast, SUA levels 7–8 mg/dl (OR: 1.12, 95% CI [1.02–1.23], P = 0.017), 8–9 mg/dl (OR: 1.18, 95% CI [1.04–1.34], P = 0.012), and ≥ 9 mg/dl (OR: 1.17, 95% CI [1.00–1.37], P = 0.049) were associated with an increased risk of AKI.
In females, SUA levels were significantly associated with 180-day (P < 0.001), one-year (P < 0.001), and overall mortality (P < 0.001). SUA < 3 and 6–7 mg/dl were linked to increased mortality (Table 4). In the multivariate logistic regression analysis examining risk factors for morbidity, SUA levels were significantly associated with composite complications (P < 0.001), CVD (P = 0.003), wound dehiscence (P = 0.005), pneumonia (P = 0.002), and AKI (P < 0.001). When analyzing the association between SUA groups and clinical outcomes, SUA levels of 3–4 were associated with a reduced risk of composite complications (OR: 0.91, 95% CI [0.84–0.98], P = 0.009) and pneumonia (OR: 0.86, 95% CI [0.73–1.00], P = 0.046), while SUA levels ≥ 8 mg/dl were significantly associated with the risk of composite complications (OR: 1.34, 95% CI [1.11–1.61], P = 0.002), CVD (OR: 2.25, 95% CI 1.18–4.29, P = 0.013), and AKI (OR: 1.41, 95% CI [1.14–1.73], P = 0.002). Supplementary Tables 3 and 4 show the adjusted associations for additional postoperative complications, including stroke, venous thromboembolism, surgical site infection, infection, sepsis, and urinary tract infection that were not statistically significant in either male or female patients.
Fig. 1 shows multivariable-adjusted spline analyses of the relationship between SUA levels and mortality at 180-day (Figs. 1A and D), one year (Figs. 1B and E), and overall (Figs. 1C and F) in male and female patients, respectively. In males, the reference SUA level for 180-day mortality was 6.17 mg/dl, with a lower-risk SUA range of 4.65–7.70 mg/dl (Fig. 1A). SUA levels outside this range significantly increased mortality risk (P < 0.001 for nonlinearity). Both sexes exhibited significant differences in survival rates across SUA groups (Supplementary Figs. 2 and 3).
Subgroup analysis according to age, ASA classification, and ESC surgical risk according to sex
In male patients, Cox regression analysis stratified by age (< 65 vs. ≥ 65 years), ASA classification (1, 2 vs. ≥ 3), ESC surgical risk (low, intermediate vs. high), and baseline renal function (eGFR < 60 vs. ≥ 60 ml/min/1.73 m2) showed that the prognostic effects of SUA levels on 180-day, one-year, and overall mortality varied significantly according to age, ASA classification, ESC surgical risk, and eGFR (all Pint < 0.001) (Supplementary Table 5). For composite complications, the prognostic effect of SUA levels differed significantly by age (Pint < 0.001) and eGFR (Pint < 0.001). In female patients, the prognostic effects of SUA levels on 30-day mortality differed significantly according to ASA classification (Pint = 0.005) only, while those for one-year mortality were significantly different according to age (Pint = 0.021), ASA classification (Pint < 0.001), ESC surgical risk (Pint < 0.001), and eGFR (Pint < 0.001). (Supplementary Table 6). In terms of composite complications, the prognostic effect of SUA levels was significantly different according to age (Pint < 0.001), ASA classification (Pint < 0.001), and eGFR (Pint < 0.001). For 30-day, 180-day, one-year, overall mortality, and composite complications, the HRs and ORs in the extreme SUA categories were generally higher in the low-risk group for both sexes.
Discussion
In this retrospective study involving 295 267 patients who underwent elective non-cardiac surgery, we identified significant associations between preoperative SUA levels—encompassing both hyperuricemia and hypouricemia—and postoperative outcomes. The mean SUA levels were 5.5 mg/dl in men and 4.2 mg/dl in women, and the lower-risk preoperative SUA range associated with reduced mortality was 5.08–7.63 mg/dl for men and 3.34–5.35 mg/dl for women. The relationship between SUA levels and mortality exhibited a nonlinear (J-shaped or U-shaped) pattern in both sexes. In men, this nonlinear association was statistically significant for 180-day, one-year, and overall mortality, while in women, it was significant only for overall mortality. Furthermore, the analysis of the relationship between SUA levels and postoperative complications showed a nonlinear pattern in composite complications, with a noticeable decrease in complication risk at certain SUA levels in both sexes. However, there were subtle differences in the types of complications observed between men and women. Additionally, the prognostic effect of SUA varied according to age, ASA, and ESC risk, as well as composite complications.
Several mechanisms have been proposed to explain the association between preoperative SUA levels and postoperative outcomes. Mechanistically, the impact of SUA on postoperative outcomes can be attributed to its dual role as both an antioxidant and a pro-oxidant, depending on its concentration. Elevated SUA levels can promote oxidative stress and inflammation, leading to vascular endothelial dysfunction, impaired nitric oxide production, HTN, and atherosclerosis [22,23]. These processes not only increase the risk of cardiovascular events [24] but may also exacerbate kidney damage and postoperative AKI through pro-inflammatory mechanisms, including chemokine production and leukocyte infiltration [25,26]. Additionally, hyperuricemia has been implicated in cancer recurrence and mortality, likely due to its role in promoting systemic inflammation and oxidative stress [27]. Conversely, hypouricemia may diminish the body’s antioxidant defenses, making cells more vulnerable to oxidative stress that can be particularly detrimental to cardiovascular, renal, and neurological health [17,28–30]. Low SUA levels have been associated with a higher risk of neurodegenerative diseases, such as Parkinson’s disease and delirium, due to reduced antioxidant protection against neuronal damage [9,31]. Furthermore, hypouricemia has been linked to malnutrition that may further contribute to adverse health outcomes by impairing overall physiological resilience and recovery [32].
Recent studies have indicated that SUA levels exhibit a J-shaped association with mortality outcomes, suggesting that both low (hypouricemia) and high (hyperuricemia) SUA levels are linked to an increased risk of adverse outcomes [15,33]. Additionally, previous research has highlighted differences in average SUA levels between males and females, indicating that the impact of SUA on postoperative outcomes may vary by sex [17,33,34]. However, comprehensive analyses investigating sex-specific differences in the effects of SUA levels on postoperative outcomes in non-cardiac surgery remain limited. This study is the first to demonstrate sex-specific associations between a modifiable biomarker (SUA) and postoperative outcomes in a large cohort of non-cardiac surgery patients. By examining the relationship between preoperative SUA levels and both short- and long-term mortality and morbidity across sexes, our study helps bridge this gap. Notably, our study holds clinical significance as it identifies a lower-risk range of preoperative SUA levels associated with postoperative mortality in both sexes, thereby providing a broader perspective for risk stratification, moving beyond reliance on precise cutoff points. A key finding of our study is the consistent impact of SUA levels on outcomes for both sexes, with subtle yet important differences.
In terms of mortality, spline analyses demonstrated a nonlinear (U-shaped) relationship in both men and women; however, the association was more pronounced in men, with significant effects observed across 180-day, one-year, and overall mortality. In contrast, in women, a significant association was noted only for overall mortality. In Cox regression analysis, both hypouricemia (< 4 mg/dl) and hyperuricemia (≥ 9 mg/dl) were significantly associated with increased mortality in men. In women, hypouricemia (< 3 mg/dl) was significantly associated with increased mortality, and although the 6–7 mg/dl range showed some significant associations, higher SUA levels (≥ 7 mg/dl) had limited or nonsignificant impact. The attenuated association between hyperuricemia and mortality in females observed in our study may be attributed to sex-specific physiological and hormonal differences. Women generally have lower baseline SUA levels due to lower muscle mass and the uricosuric effect of estrogen that enhances renal uric acid excretion [35]. As a result, even when SUA levels are elevated, they may not reach the pathophysiological threshold required to exert the same adverse effects observed in males. This may explain why hyperuricemia was less strongly associated with mortality in females compared to males.
In terms of postoperative complications, SUA levels exhibited notable sex-specific differences. High SUA levels were associated with composite complications and AKI in both sexes. However, males were more likely to experience bleeding complications, whereas females had higher risks of CVD, wound dehiscence, and pneumonia. This is consistent with findings from a large-scale population-based study in Koreans that reported an association between SUA levels and cardiovascular mortality observed only in women [33]. Low SUA levels were linked to a reduction in composite complications in both sexes, likely due to lower oxidative stress. Nonetheless, extremely low SUA levels may impair antioxidant defenses, potentially increasing risks in certain subgroups of complications [15].
In our study, specific SUA ranges (4–6 mg/dl in males and 3–4 mg/dl in females) demonstrated a protective effect against composite and certain complications in both sexes. These findings raise the question of whether modifying preoperative uric acid levels could improve surgical outcomes. While this remains to be proven, a small pilot study did find that pharmacologically lowering uric acid (with rasburicase) was associated with reduced AKI in high-risk patients [36]. However, the long-term effects and safety of SUA modulation in surgical populations remain unclear. Further research is needed to determine whether targeted SUA management could serve as a viable perioperative intervention to optimize postoperative outcomes.
Our subgroup analyses suggest that the management of SUA levels should be tailored to individual patient risk factors as well as sex. In low-risk groups, such as younger patients (< 65 years), those classified as ASA 1–2, or those categorized as ESC low/intermediate surgical risk, extreme SUA levels were consistently associated with an increased risk of mortality and complication compared to high-risk groups. This trend may indicate that extreme SUA levels in low-risk groups are more likely to disrupt the metabolic balance, triggering acute metabolic stress or oxidative stress. These extreme SUA levels may also reflect the severity of underlying diseases or acute metabolic crises. In contrast, high-risk groups exhibited lower mortality and complication risks compared to low-risk groups that may indicate a more chronic and adaptive response to SUA dysregulation. Similarly, patients with preserved renal function (eGFR ≥ 60 ml/min/1.73 m²) demonstrated stronger and more consistent associations between SUA and adverse outcomes than those with impaired renal function. This suggests that renal function may not only confound the relationship between SUA and surgical outcomes but also modify its prognostic impact. These results reinforce the importance of considering renal function status when interpreting SUA in the perioperative setting, as the prognostic value of SUA may vary according to the degree of underlying renal reserve. Taken together, these subgroup findings emphasize the significant interaction between patient risk status (low-risk vs. high-risk) and SUA levels in determining surgical outcomes.
Our study has several limitations. First, due to its retrospective design, there may be unmeasured or missing outcomes that could have influenced the results. Additionally, relying on electronic medical records for data collection may have introduced bias stemming from incomplete or inaccurately recorded information. In particular, the overall complication rates observed in our cohort were lower than those reported in previous studies that may reflect the exclusion of emergency surgeries, a predominance of lower-risk patients, or potential under-detection of complications inherent to EMR-based retrospective analysis. Second, this single-center study focused on an ethnically homogeneous group of primarily Korean patients. Third, patients taking medications that affect SUA levels, such as urate-lowering therapies (e.g., allopurinol, febuxostat), were not excluded from this study. To minimize this limitation, we adjusted the multivariate analysis for patients with conditions like gout or kidney stones. Fourth, while we demonstrated the association between preoperative SUA and postoperative surgical outcomes, the study design does not allow us to infer a direct causal relationship. Lastly, as SUA values were measured at a single point in time before surgery as a baseline, the dynamic changes in SUA that may affect postoperative outcomes may not have been fully captured.
In conclusion, our study highlights the complex, nonlinear relationship between SUA levels and postoperative outcomes, revealing sex-specific differences in lower-risk SUA range and their impact on mortality and complications. These findings underscore the necessity for personalized perioperative management strategies that consider SUA levels, sex, and patient risk profiles to enhance surgical outcomes.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Supplement1
Print
