Abstract
Objective:
To develop and validate a nomogram for predicting the occurrence of systemic inflammatory response syndrome (SIRS) following percutaneous nephrolithotomy (PCNL), aiming to enhance clinical decision-making and treatment planning.
Methods:
Clinical data of 1,047 patients undergoing PCNL at a single-center hospital between 2017 and 2023 were retrospectively analyzed. Independent risk factors influencing SIRS occurrence were identified through multi-variable logistic regression analysis, and a predictive model was constructed. The model’s accuracy and reliability were evaluated through internal training and validation set.
Results:
Multi-variable regression analysis identified six key predictive factors: gender, diabetes, urine culture results, stone surface, staghorn stones, and operative time, leading to the establishment of a nomogram predictive model. Internal validation and validation set data demonstrated the model’s high predictive accuracy and reliability, with areas under the receiver operating characteristic curve of 0.718 and 0.723, respectively.
Conclusions:
A nomogram predictive model for assessing SIRS following PCNL was successfully developed and validated. This model provides clinicians with a valuable tool for personalized treatment planning and implementing preventive measures.
Introduction
Urolithiasis is a prevalent disorder within the urinary system, with its global incidence steadily rising. 1 Prevalence rates stand at 8.8% in North America and 6.5% in China, substantially impacting patients’ quality of life and overall health. 2 With continuous advancements in medical technology, percutaneous nephrolithotomy (PCNL) has emerged as a preferred method for treating such conditions. 3 Compared with traditional open operation, PCNL offers reduced trauma and quicker recovery times, leading to its widespread adoption in clinical practice. 4 Despite substantial advancements in enhancing treatment efficacy through PCNL, the occurrence of post-operative complications remains a critical determinant of surgical success and patient quality of life. 5
Systemic inflammatory response syndrome (SIRS) represents a severe post-operative complication that may arise following PCNL. 6 Characterized by an excessive systemic inflammatory response, it can lead to tissue and organ damage, circulatory disturbances, and even urosepsis, with potential life-threatening consequences.7,8 The high incidence of SIRS poses substantial challenges to patient recovery and treatment, underscoring the crucial importance of timely prediction and intervention. 9
Currently, prediction and intervention for post-operative SIRS rely primarily on clinicians’ subjective judgment and conventional examinations, which often struggle to accurately assess post-operative risks for patients. 10 Thus, it is critically important to create a straightforward, precise predictive model that allows healthcare providers to quickly pinpoint patients at high risk and adopt tailored preventive tactics. Nomograms, renowned for their intuitive and user-friendly nature, have showcased remarkable predictive capabilities across diverse medical domains. 11 Despite recent strides in developing nomogram models to predict the risk of SIRS after PCNL, several studies suffer from limitations such as small sample sizes or absence of rigorous internal or external validation.12–14 These constraints curtail the clinical utility and generalizability of these models.
In light of this, our study aims to retrospectively analyze data from 1,047 patients who underwent PCNL treatment at our hospital between 2017 and 2023. Utilizing R software, we will allocate these data into training (733 cases) and validation (314 cases) sets at a ratio of 7:3. Through multi-variable logistic regression analysis conducted on the training set, we aim to identify independent risk factors influencing SIRS following PCNL and construct a nomogram predictive model. Subsequently, we will evaluate the accuracy and reliability of this model through training and validation set. Through this study, we anticipate providing clinicians with a straightforward and dependable predictive tool. This will enhance the early recognition of post-operative SIRS, enabling timely and effective interventions and ultimately improving patient treatment outcomes and quality of life.
Materials and Methods
Study population
A retrospective study was conducted to collect data from patients who underwent PCNL at our institution between September 2017 and August 2023. Inclusion criteria comprised: (1) age 18 years or older; (2) diagnosis of urolithiasis confirmed by computed tomography or kidney-ureter-bladder radiograph and subsequent treatment with PCNL; and (3) absence of severe infection or inflammatory diseases preoperatively. Exclusion criteria included: (1) incomplete clinical records; (2) preoperative presence of SIRS or severe infection; and (3) post-operative complications that interfere with the assessment of SIRS, such as severe bleeding, critical post-operative infections, intensive care unit admissions for severe conditions, substantial neurological complications, or the use of certain strong medications.
Data selection
Data extracted from medical records included gender, age, body mass index (BMI), the presence of diabetes mellitus, hyperlipidemia, hypertension complications, smoking history, history of urinary tract infections (UTIs), positive urine culture, and morphology of calculi such as staghorn stones. In addition, operation-related parameters including operative time, intra-operative blood loss, surface area, location, and number of calculi were collected. Calculation of stone surface was determined using the formula: stone surface area (mm2) = L * W * π * 0.25, where L and W represent the length and width of the stone, respectively, obtained through imaging studies. 15
Definition of post-operative SIRS
SIRS was defined as the presence of at least two or more of the following clinical criteria: abnormal body temperature (<36°C or >38°C), tachycardia (heart rate >90 beats per minute), abnormal respiratory rate (>20 breaths per minute or PaCO2 < 32 mmHg), and abnormal white blood cell count (>12,000/mm3 or <4,000/mm3 or >10% immature bands).16,17
Statistical analyses
All statistical analyses were conducted using SPSS version 26.0 and R software version 4.2.2. Patients were randomly assigned to either the training or validation groups in a 7:3 ratio using R software to ensure model robustness. Descriptive statistics summarized parameters, with continuous ones presented as mean ± standard deviation and categorical ones as frequency (percentage). Uni-variable analyses, including chi-square tests and t tests, identified potential risk factors associated with SIRS. Significant parameters (p < 0.05) were included in multi-variable logistic regression analysis to determine independent predictors of SIRS. A nomogram prediction model was then developed based on the multi-variable analysis results using the “rms” package in R software. Harrell’s concordance index (C-index) and the area under the curve (AUC) assessed the nomogram’s predictive accuracy and discriminative ability. Calibration curves evaluated the agreement between predicted and observed outcomes. Model robustness was confirmed through 1,000 bootstrapping resamples. Decision curve analysis (DCA) assessed the clinical utility and net benefit of the nomogram using the “rmda” package.
Ethics
Ethical approval was obtained from the Ethics Committee of The Affiliated Hospital of Jiangxi University of Traditional Chinese Medicine. Permission to conduct the study was granted by the hospital. The requirement for informed consent was waived because of the retrospective nature of the study and the use of anonymized patient data.
Results
Clinicopathologic characteristics
From 2016 to 2023, patients with urolithiasis were screened from a single-center database for inclusion in this study. After applying inclusion and exclusion criteria, a total of 1,047 patients were enrolled (Fig. 1). Using R software, these patients were randomly assigned to training (733 cases) and validation groups (314 cases) at a 7:3 ratio. No significant differences in clinical characteristics were observed between the training and validation groups (p > 0.05; Table 1).
Flow diagram of study design. PCNL = percutaneous nephrolithotomy; SIRS = systemic inflammatory response syndrome.
Patient Characteristics for the 1,047 Study Participants
BMI = body mass index; UTI = urinary tract infection; SIRS = systemic inflammatory response syndrome.
In addition, the 1,047 participants were stratified into the SIRS group (136 cases) and the non-SIRS group (911 cases), with a post-operative incidence rate of SIRS at 13.0%. Analytical findings revealed no significant differences between the SIRS and non-SIRS groups concerning age, BMI, smoking history, hypertension, hyperlipidemia, history of UTI, stone location, stone number, and post-operative estimated blood loss (all p > 0.05). However, significant differences were observed between the SIRS and non-SIRS groups in terms of gender, diabetes, urine culture positivity, stone surface area, staghorn stone presence, and operative time. Specifically, the proportion of female patients in the SIRS group (54.4%) was slightly greater than that in the non-SIRS group (40.2%), with gender exerting a significant influence on SIRS occurrence (p = 0.002). Furthermore, the incidence rate of diabetes in the SIRS group (20.6%) was significantly greater than that in the non-SIRS group (11.5%), with statistical significance (p = 0.003). Urine culture positivity in the SIRS group (18.4%) was also significantly greater than that in the non-SIRS group (9.5%; p = 0.002). Stone surface ≥ 1,000 mm2 and the presence of staghorn stone were significantly associated with SIRS (p < 0.001). In addition, the proportion of patients with operative time exceeding 2 h in the SIRS group (39.0%) was significantly greater than that in the non-SIRS group (18.2%), with p < 0.001.
In summary, factors such as gender, diabetes, urine culture positivity, stone surface, the presence of staghorn stone, and operative time are significantly associated with the post-operative SIRS.
Independent predictive factors for SIRS
The logistic regression analysis in Table 2 identified significant predictors for SIRS occurrence. Notably, female gender (odds ratio [OR] = 1.85, 95% confidence interval [CI] 1.26–2.72, p = 0.002), the presence of diabetes (OR = 2.00, 95% CI 1.20–3.33, p = 0.007), positive urine culture (OR = 2.02, 95% CI 1.19–3.42, p = 0.009), stone surface ≥1,000 mm2 (OR = 1.96, 95% CI 1.33–2.86, p < 0.001), the presence of staghorn stone (OR = 1.89, 95% CI 1.22–2.85, p = 0.004), and operative time ≥2 h (OR = 2.67, 95% CI 1.91–3.72, p < 0.001) were significantly associated with SIRS risk. These findings underscore the importance of these factors in predicting post-operative SIRS.
Logistic Regression Analysis of the Predictors for SIRS
CI = confidence interval; SIRS = systemic inflammatory response syndrome.
Development and validation of the nomogram
A nomogram based on recognition of independent predictive factors was constructed to predict the risk of SIRS after PCNL. As depicted in Figure 2, this nomogram incorporates six key parameters: gender, diabetes, urine culture, stone surface, staghorn stones, and operative time. Each variable is assigned specific scores, allocating different points based on their respective levels: for instance, gender (male: 0 points; female: 61.25 points), diabetes (absent: 0 points; present: 70 points), urine culture (negative: 0 points; positive: 61 points), stone surface (<1,000 mm2: 0 points; ≥ 1,000 mm2: 58.75 points), staghorn stones (no: 0 points; yes: 67.5 points), and operative time (<2 h: 0 points; ≥2 h: 100 points). By aggregating these individual scores, a total score is calculated, providing a practical tool to predict the risk of SIRS following PCNL and aiding clinical decision-making for prudent treatment choices.
The graph showed nomogram for predicting SIRS. SIRS = systemic inflammatory response syndrome.
The overall performance of the nomogram was assessed, revealing a C-index of 0.718 for the training set and 0.723 for the validation set, indicative of the robust discriminative capacity of the predictive model. Furthermore, receiver operating characteristic curve analysis displayed an AUC of 0.718 (95% CI 0.672–0.764) for the training set and 0.723 (95% CI: 0.649–0.797) for the validation set (Supplementary Fig. S1A and B), affirming the model’s efficacy in discriminating outcomes. Calibration curves for both sets exhibited strong concordance between the predicted probabilities by the nomogram and the observed outcomes (Supplementary Figure S1C and D). DCA delineated threshold probabilities for the predictive model spanning from 6% to 71% in the training set and from 5% to 48% in the validation set, underscoring its clinical applicability across diverse risk thresholds (Supplementary Figure S1E and F).
Discussion
PCNL is one of the mainstays in the management of urinary stone disease. 18 However, post-operative complications, particularly SIRS, remain crucial determinants affecting surgical outcomes and patient prognoses. 19 Thus, the timely and accurate assessment of post-operative SIRS risk is paramount for clinicians.
In the present study, a post-PCNL SIRS incidence rate of 13.0% was observed, in line with previous findings.12,13 The prompt detection of biomarkers holds substantial promise for the early diagnosis of SIRS. Previous investigations have underscored the importance of urine culture positivity, a history of stone operation, stone size, stone complexity, and operative time as primary risk factors associated with SIRS.20–22 Although some studies have delved into biomarkers, further exploration and validation are deemed necessary.
Several predictive factors associated with SIRS following PCNL were identified in this study. First, the influence of gender on SIRS occurrence was noted. The findings revealed a heightened susceptibility to SIRS among female patients, consistent with prior research. 19 Gender disparities may be attributed to the shorter female urethra, rendering them more vulnerable to infection-induced inflammation of the urinary tract.23,24 This underscores the imperative for intensified infection prevention measures preoperatively or intra-operatively in female patients. Second, diabetes emerged as another significant predictive factor. Diabetic patients exhibited a greater propensity for SIRS, likely attributable to compromised immune function and impaired tissue repair associated with diabetes.25,26 This underscores the importance of pre-operative immune status assessment, alongside intensified pre-operative glycemic control and infection prevention measures, to mitigate SIRS risk among diabetic patients undergoing PCNL. Third, urinary culture results also demonstrated a significant correlation with SIRS occurrence. Positive urine culture findings closely correlated with SIRS, highlighting UTI as a prominent precipitant. 27 In clinical practice, patients with pre-operative positive urine cultures necessitate intensified anti-infective therapy and stringent aseptic measures during operation to mitigate infection risks. Fourth, stone surface area and staghorn stones were identified as pivotal factors influencing SIRS occurrence. Patients with larger stone surface areas and staghorn stones exhibited a heightened propensity for SIRS, likely because of severe urinary obstruction and infection, triggering more robust inflammatory responses.12,28 Consequently, comprehensive assessment of stone size and morphology is imperative in surgical planning and pre-operative evaluation, guiding appropriate surgical tactics and preventive measures to mitigate SIRS incidence. Finally, operative duration emerged as a significant determinant of SIRS occurrence. Our results indicated a greater likelihood of SIRS among patients with operative durations exceeding 2 h. Prolonged operations may precipitate tissue ischemia-reperfusion injury and heightened infection risks, prompting inflammatory responses. 29 Hence, minimizing operative duration is paramount, with closer intra-operative monitoring recommended for complex cases to mitigate complication risks.
Although previous studies have attempted to develop different nomograms to predict the risk of SIRS after PCNL, the majority either had small sample sizes or lacked internal or external data validation, limiting their applicability and generalizability in clinical practice.12–14 In contrast, our study, based on data from 1,047 patients undergoing PCNL, was constructed and validated. Through analysis and selection of multiple predictive factors, we identified gender, diabetes, urine culture, stone surface, staghorn stones, and operative time as six key predictive factors, integrating them into a simple yet effective nomogram. The model demonstrated high accuracy and reliability in predicting post-PCNL SIRS occurrence following validation with internal training and validation sets, offering critical decision support for clinicians.
Although our study has progressed, there are limitations to consider. First, it is a single-center retrospective study, which may introduce biases in patient selection and data collection. Despite our efforts to address these biases, complete elimination is challenging. Future multi-center, prospective studies could provide more robust validation. Second, although our model performed well in validation, there are still limitations. It may not account for all relevant predictive factors and might benefit from finer data resolution. Thus, further research could explore additional factors and refine the model. In particular, the absence of A1C values in our dataset prevented us from assessing the impact of diabetes control on SIRS risk. Future studies should consider incorporating A1C concentrations to determine whether uncontrolled diabetes contributes more significantly to SIRS development compared with a diabetes diagnosis alone. In addition, although our sample size is considerable, it remains limited, affecting the model’s stability and generalizability. Increasing the sample size could enhance validation. Furthermore, regional variations may exist, necessitating validation across diverse populations. Lastly, although our nomogram shows promise for clinical application, it requires clinical expertise. Although it aids prediction, treatment decisions should consider individual patient factors. Hence, comprehensive assessment and decision-making are essential in clinical practice.
Despite these limitations, our study provides new insights into the management of SIRS following PCNL. Utilizing our nomogram model, clinicians can offer more accurate risk assessment and treatment recommendations to patients. Future research should focus on refining and validating this model for broader clinical applicability across different populations.
In summary, the developed nomogram represents a valuable tool for predicting the risk of SIRS following PCNL. Through comprehensive analysis of key predictive factors, this model facilitates personalized patient management and treatment decision-making. Reducing post-operative SIRS can improve patient quality of life and alleviate the burden on healthcare systems. We hope that this model will receive widespread adoption and enhancement in various clinical settings.
Conclusion
This study successfully developed and validated a nomogram for predicting the risk of SIRS following PCNL. The model integrates several important predictive factors including gender, diabetes, urine culture results, stone surface, staghorn stones, and operative time, offering substantial clinical value. It enables clinicians to provide more precise risk assessment and treatment recommendations to patients.
Footnotes
Authors’ Contributions
Conceptualization: C.F., Q.J., J.T., and Z.J.W. Methodology: C.F., Q.J., J.T.T., and Z.J.W. Software: C.F. and Q.J. Validation: C.F. and Q.H.J. Formal analysis: C.F., Q.H.J., J.T.T., and Z.J.W. Investigation: C.F., Q.H.J., and J.T.T. Data curation: C.F. Writing—original draft: C.F., Q.H.J., J.T.T., and Z.J.W. Writing—review and editing: C.F. and Q.H.J. Visualization: C.F. and Q.H.J. Resources: J.T.T. and Z.J.W. Supervision: J.T.T. and Z.J.W.
Author Disclosure Statement
The authors report no conflicts of interest.
Funding Information
No funding was received for this article.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
