Development and external validation of a novel score for predicting postoperative 30‑day mortality in tumor craniotomy patients: A cross‑sectional diagnostic study
- Authors:
- Published online on: March 12, 2024 https://doi.org/10.3892/ol.2024.14338
- Article Number: 205
-
Copyright: © Liu et al. This is an open access article distributed under the terms of Creative Commons Attribution License.
Abstract
Introduction
Craniotomy is a basic surgical procedure for managing most patients with brain tumors. However, craniotomies for intracranial tumors are associated with significant and numerous risks of postoperative complications, including death (1–3). Postoperative 30-day mortality, which is also known as 30-day postoperative mortality, is widely used to assess the short-term outcomes of patients undergoing various surgeries (4,5). It is also used to evaluate the effectiveness of access to and safety of anesthesia and surgery (6). Postoperative 30-day mortality was shown to be 5.03% in an American study of 16,280 patients who underwent craniotomy (7). Another study of craniotomy patients treated from 2008–2010 at multiple centers in England reported a range of mortality rates from 0.95 to 8.62% (8). Therefore, obtaining accurate individualized preoperative risk predictions of short-term outcomes is important for clinical decision-making and further management.
Numerous predictive scoring systems for the severity of illness or prognosis, such as the Glasgow Coma Scale for traumatic brain injury (9), the Hunt and Hess scale for aneurysmal subarachnoid hemorrhage (10) and the Unified Parkinson's Disease Rating Scale for Parkinson's disease, have been widely used in neurology (11). Accordingly, previous studies have attempted to construct diagnostic or prognostic prediction models for patients with various intracranial tumors, including gliomas (12–15), meningiomas (16,17), brain metastases (18–20), clival chordomas (21) and medulloblastomas (22); in addition, the clinical value of these nomograms has been emphasized. This research has focused mainly on a single disease, and a small number of studies have focused on the risk prediction of prognosis after craniotomy in patients with brain tumors (12–14,18–20). Several preoperative risk factors for postoperative pneumonia after craniotomy have been identified based on an American database (2005–2017) (23). However, to the best of our knowledge, neither nomograms nor preoperative scoring systems have been reported to evaluate and predict 30-day mortality risk after brain tumor craniotomy. In the present study, a novel scoring system for predicting postoperative 30-day mortality was developed in 18,642 craniotomy patients. It is anticipated that the mortality risk prediction model will help clinicians (particularly neurosurgeons), patients and their families assess postoperative 30-day mortality and choose related and positive interventions to prevent or reduce mortality.
Patients and methods
Study design and population
A retrospective analysis of 18,642 participants with brain tumors who underwent craniotomy between 2012 and 2015 was performed; the information regarding these patients was retrieved from the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) database (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7498000/, S1 Data). The ACS NSQIP is a validated, prospectively collected, publicly available, peer-controlled database of a random sample of outpatients and inpatients undergoing nontrauma surgery at ~400 community and academic hospitals across the US. The identities of the patients were encrypted as nontraceable codes to ensure participant privacy. Variables at baseline were included as screening variables in the prediction model in the present study. The dependent variable was postoperative 30-day mortality (dichotomous variable: 0=nonpostoperative 30-day mortality; 1=postoperative 30-day mortality).
Data source
Zhang et al (24) previously published an article titled ‘Sepsis and septic shock after craniotomy: Predicting a significant patient safety and quality outcome measure’ and uploaded the original data to the ACS NSQIP database. The uploaded data are available for use in secondary analyses without infringement on the authors' rights and the copyright statement.
Variables
The following variables were extracted for the present study according to the previous literature and our clinical experience: i) Continuous variables, including body height, body weight and indicators of preoperative blood test results [hematocrit (HCT), blood urea nitrogen (BUN), white blood cell (WBC) count, creatine (Cr) and platelet (PLT) count], and ii) categorical variables, including sex, ethnicity, age range, diabetes status, smoking status, year of operation, dyspnea, functional health status, ventilator dependence, severe chronic obstructive pulmonary disease (COPD), congestive heart failure (CHF), hypertension, renal failure, preoperation transfusions, dialysis, disseminated cancer, preoperative systemic sepsis, open wound infection, steroid use for chronic conditions, >10% loss of body weight in the last 6 months, bleeding disorders, emergency cases, wound classification and American Society of Anesthesiologists (ASA) physical status classification. More elaborate details were presented in the original study (24). The body mass index (BMI) was calculated as weight in kilograms divided by height in meters squared (kg/m2).
Handling of missing baseline variables
The number of participants with missing BMI (weight and height), functional health status, Na, BUN, Cr, WBC, HCT, PLT and ASA data was 730 (3.92%), 90 (0.48%), 798 (4.28%), 1,532 (8.22%), 709 (3.8%), 592 (3.18%), 440 (2.36%), 579 (3.11%) and 166 (0.89%), respectively. Multiple imputation techniques are widely accepted as appropriate methods for handling missing data (25). This method was used to input missing values for the extracted variables in the present study. The imputation model included BMI; functional health status; Na, BUN, and Cr levels; WBC count; HCT level; PLT count; and ASA classification. Missing data analysis procedures used missing-at-random assumptions (26).
Outcome measures
The primary outcome variable was postoperative 30-day mortality. The NSQIP was used to track mortality for the first 30 postoperative days (24).
Statistical analysis
A training dataset (patients who underwent craniotomy in 2012 and 2013) and an external validation dataset (those who underwent craniotomy in 2014 and 2015) were generated from the initial study population. The training dataset was used to establish the model and the external validation dataset was used for independent evaluation of the preliminary model's performance.
Baseline characteristics are expressed as the mean ± standard deviation (normal distribution) or the median (interquartile range) (skewed distribution) for continuous variables and as the frequency and percentage for categorical variables. Two-samples t-tests were applied to analyze differences between the training and validation cohorts for normally distributed continuous variables. Wilcoxon rank-sum tests were used for nonnormally distributed continuous variables, and chi-square test or Fisher's exact test was used for categorical variables. The baseline characteristics of the training and validation cohorts stratified were also presented with stratification by incident 30-day mortality. Univariate and multivariate analyses were also performed to identify potential risk factors of 30-day postoperative mortality after craniotomy for brain tumors.
To construct a reliable and simple risk prediction model, two rounds of variable screening were conducted. The least absolute shrinkage and selection operator (LASSO) method is frequently used for domains with very large datasets and is suitable for the reduction of high-dimensional data (27). This dataset was used to select the most useful predictive candidates from the training dataset. Candidates with nonzero coefficients in the LASSO regression model were selected (28). A second screening round was performed based on the LASSO model's identified variables. First, all of the risk factors were applied to construct a full logistic regression model. Second, a backward step-down selection process was conducted according to the Akaike information criterion to establish a parsimonious model (a stepwise logistic proportional hazards model) (29). Third, according to the multivariable fractional polynomial (MFP) algorithm, an iterative approach was used to determine the significant variables and functional form via backward elimination to establish a stable model (MFP model) in the real world (30). Considering that there were fewer variables in the stepwise model and that the predictive performance was relatively good, the stepwise model was selected for further analysis.
To evaluate and compare the discriminatory power of these prediction models, the receiver operating characteristic (ROC) curve was plotted and the area under the ROC curve (AUC) with 95% confidence intervals (CIs) was calculated for the training dataset and validation dataset. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), positive likelihood ratio (PLR), negative likelihood ratio (NLR) and diagnostic odds ratio (DOR) of the stepwise model, which were calculated according to standard definitions, were simultaneously presented. A prediction formula was obtained from the stepwise logistic proportional hazards model. The nomogram was based on proportionally converting each regression coefficient in the multivariate logistic regression to a 0- to 100-point scale (31). The effect of the variable with the highest β coefficient (absolute value) was assigned 100 points. The points were added across independent variables to derive total points, which were converted to predicted probabilities of postoperative 30-day mortality. The nomogram score was a numeric value representing the prediction model score of the individual patient. The sensitivity and specificity for predicting 30-day mortality were different at different cutoff values of the nomogram scores. In addition, a calibration plot for the probability of 30-day mortality was generated to assess the accuracy of the nomogram (32).
The associated risk factors for 30-day mortality in the stepwise model were also categorized according to clinical cutoff points to create the score model of 30-day mortality. These risk factors, which were treated as categorical variables, were included in the stepwise logistic proportional hazards model and a new β coefficient was derived. The scoring system was developed based on regression coefficients multiplied by 2 and rounded to the nearest integer to derive the weights of the scores (33). This scoring system was subsequently presented as a questionnaire form that can be easily used by health personnel in primary care. The total score was divided into four risk categories: Low, intermediate, high and extremely high risk categories. The performance of our risk score model for predicting postoperative 30-day mortality was also tested by analyzing the performance of each risk factor in the model and its optimal cutoff for predicting postoperative 30-day mortality based on ROC curves. All of the results are reported according to the TRIPOD statement (34).
All of the analyses were performed with the statistical software packages R (http://www.R-project.org; The R Foundation) and EmpowerStats (http://www.empowerstats.com; X&Y Solutions, Inc.). All of the tests were 2-sided, with P<0.05 considered to indicate statistical significance.
Results
Baseline characteristics of patients
The current study included 18,642 adult participants (47.40% of whom were men) (Table SI). The age distributions were 16.40% (18–40 years), 41.53% (41–60 years), 38.80% (61–80 years) and 3.27% (>81 years). The mean BMI was 28.69±6.72 kg/m2, the mean Na concentration was 138.62±3.22 mmol/l, the mean BUN concentration was 17.39±8.31 mg/dl, the mean Cr concentration was 0.87±0.45 mg/dl, the mean WBC count was 9.50±4.48×109/l, the mean HCT level was 40.35±4.81%, and the mean PLT count was 243.4±76.90×109/l. The postoperative 30-day mortality of the included participants was 2.46% (458/18,642).
Characteristics of patients in different groups
Table I shows the basic demographic, anthropological and clinical information for the eligible participants. The participants were assigned to two groups based on the year of surgery: The training dataset (2012–2013) and the validation dataset (2014–2015). For numerous baseline characteristics, although the differences between the training cohort and the validation cohort were statistically significant due to the large sample size (P<0.05), they were not clinically significant.
Table II shows the baseline characteristics of patients with nonpostoperative 30-day mortality and postoperative 30-day mortality in the training and validation datasets. The participants with postoperative 30-day mortality had higher BUN levels and WBC counts in the training and validation cohorts (all P<0.01). By contrast, the participants who died within 30 days postoperatively had lower Na concentrations, HCT levels and PLT counts (all P<0.05).
 | Table II.Baseline characteristics for the training and validation cohorts by incident 30-day mortality. |
Univariate and multivariate analyses
The results of the univariate and multivariate analyses using a binary logistic regression model are presented in Table SII. The univariate analysis showed that female sex (OR=0.649), age range (41–60 years) (OR=2.682), age range (61–80 years) (OR=4.940), age (>81 years) (OR=14.902), BMI (OR=0.985), diabetes (noninsulin-dependent) (OR=1.552), diabetes (insulin-dependent) (OR=2.618), dyspnea (moderate exertion) (OR=2.333), dyspnea (moderate exertion) (OR=2.333), functional health status (partially dependent) (OR=4.032), functional health status (totally dependent) (OR=7.211), ventilator dependence (OR=4.527), severe COPD (OR=2.525), CHF (OR=7.270), hypertension (OR=2.292), renal failure (OR=10.893), dialysis (OR=6.580), disseminated cancer (OR=2.913), open wound infection (OR=5.041), steroid use for chronic conditions (OR=2.330), >10% body weight loss in last 6 months (OR=4.255), bleeding disorders (OR=2.307), preoperative transfusions (OR=5.860), SIRS (OR=2.186), sepsis (OR=13.470), septic shock (OR=9.354), levels of Na (OR=0.910), BUN (OR=1.043), Cr (OR=1.240), WBC count (OR=1.074), HCT level (OR=0.923), PLT count (OR=0.998), emergency cases (OR=2.875) and wound classification (dirty/infected) (OR=5.506) were associated with postoperative 30-day mortality (all P<0.05).
The multivariate analysis demonstrated that female sex (OR=0.713), age range (41–60 years) (OR=1.927), age range (61–80 years) (OR=2.573), age (>81 years) (OR=6.680), functional health status (partially dependent) (OR=2.152), functional health status (totally dependent) (OR=2.982), ventilator-dependence (OR=2.402), hypertension (OR=1.374), disseminated cancer (OR=1.791), open wound infection (OR=2.297), steroid use for chronic conditions (OR=1.619), >10% body weight loss in the last 6 months (OR=2.067), sepsis (OR=3.563), Na level (OR=0.971), BUN level (OR=1.019), WBC count (OR=1.049), HCT level (OR=0.959), PLT count (OR=0.998), emergency cases (OR=2.267) and wound classification (dirty/infected) (OR=3.228) were associated with postoperative 30-day mortality (all P<0.05).
Candidate selection through LASSO regression
Of the clinical features, 30 indicators [BMI and preoperative blood test results (HCT, BUN, WBC, Cr, PLT), sex, ethnicity, age ranges, diabetes status, smoking status, dyspnea, functional health status, ventilator dependence, severe COPD, CHF, hypertension, renal failure, dialysis, disseminated cancer, open wound infections, steroid use for chronic conditions, >10% loss of body weight in the last 6 months, bleeding disorders, preoperative transfusions, preoperative systemic sepsis, emergency cases, wound classification and ASA physical status classification)] were reduced to 6 potential predictors based on 7,800 participants in the training dataset (Fig. 1A and B) with nonzero coefficients in the LASSO regression model. These potential predictors included the preoperative WBC count, BUN level, HCT level, age range, functional health status and disseminated cancer.
Identification of risk factors
A total of three prediction models were further established based on the predictors chosen by the LASSO regression model, namely the MFP model, the full logistic proportional hazards model and the stepwise logistic regression model. In the training cohort, the AUC values of the MFP model, full model and stepwise model were 0.7983, 0.7949 and 0.7949, respectively. In the validation cohort, the corresponding AUC values of these models were 0.7423, 0.7382 and 0.7382, respectively (Fig. S1A and B). The AUCs of the three models were relatively close. Given that the stepwise model incorporated fewer risk factors, it was simpler than the MFP and full models. In addition, the stepwise model could predict the risk of postoperative 30-day mortality relatively well. Therefore, the stepwise model was selected as the optimal risk prediction model for postoperative 30-day mortality. As indicated in Table III, 6 variables were selected according to the stepwise model: WBC count (OR=1.0710, 95% CI=1.0420–1.1009), age range (41–60 years) (OR=1.7108, 95% CI=0.8032–3.6439), age range (61–80 years) (OR=2.8297, 95% CI=1.3510–5.9270), age (>81 years) (OR=8.2427, 95% CI=3.5937–18.9056), BUN (OR=1.020, 95% CI=1.008–1.031), HCT (OR=0.945, 95% CI=0.919–0.972), functional health status (partially dependent) (OR=3.0521, 95% CI=1.9820–4.7000), functional health status (totally dependent) (OR=2.9286, 95% CI=0.9864–8.6944) and disseminated cancer status (OR=2.8180, 95% CI=2.0631–3.8490). The results showed that 5 variables (excluding the level of HCT) were positively associated with postoperative 30-day mortality.
| Table III.Variables selected using the stepwise logistic proportional hazards model in the training dataset. |
The ability of each risk factor to predict postoperative 30-day mortality was evaluated in the training and validation cohorts (Tables SIII and SIV; Fig. S2A and B). Tables SIII and SIV indicate that each risk predictor showed high accuracy in our nomogram.
Development of the nomogram
A corresponding nomogram was further constructed to provide a quantitative and simple tool for predicting the risk of postoperative 30-day mortality by using the preoperative WBC count, HCT level, BUN level, age ranges, functional health status and disseminated cancer incidence (Fig. 2). Each variable in the nomogram was assigned a specific point value and the points for each variable were summed to obtain the total points, which were used to determine the probability of postoperative 30-day mortality. The algorithm for determining the risk of postoperative 30-day mortality in the stepwise model was as follows: Log (Y)=−3.90696+0.06862 × WBC (×109/l) + 0.53696 × [age range (41–60)] (years) + 1.04019 × [age range (61–80)] (years) +2.10932 × [age range (>81)] (years) +1.11582 × (functional health status, partially dependent) +1.07452 × (functional health status, totally dependent) +1.03602 × (disseminated cancer) + 0.01967 × BUN (mg/dl)-0.05668 × HCT (%). Probability of 30-day mortality=1/{1+e[-log(Y)]}.
Predictive performance of the nomogram
DiscriminationIn the training cohort and the validation cohort, the AUCs of the nomogram were 0.7949 (95% CI=0.7644–0.8255) and 0.7382 (95% CI=0.7091–0.7674), respectively (Table IV, Fig. 3). At the best threshold, the sensitivity was 71.35 and 66.67% and the specificity was 74.96 and 69.67% for the training and validation cohorts, respectively. Of note, both the training and validation cohorts had relatively high NPVs.
Model accuracy evaluation
It was also evaluated how close the predicted postoperative 30-day mortality was to the observed postoperative 30-day mortality risk for the nomogram in the training and validation cohorts. The calibration for the probability of postoperative 30-day mortality showed excellent agreement between the predicted possibility and the actual observation in both the training and validation sets (Fig. 4). These results demonstrated that the nomogram was able to accurately predict postoperative 30-day mortality in an American population.
Risk score model of postoperative 30-day mortality
Selected continuous variables (BUN, WBC and HCT) were converted into categorical variables according to the best threshold (Table V). Score points were assigned to each risk factor by using the model parameter estimates, after which the values were multiplied by 2 and rounded to the nearest integer. The logistic estimates for the risk variables, corresponding score points and the contributed AUC for each variable are provided in Table VI.
The resulting 30-day mortality scores ranged between a minimum of −1.5 and a maximum of 9 points and were divided into four groups according to the quartile of the total risk score as follows: Low risk (−1.5 to −1), moderate risk (−0.5 to 0.5), high risk (1 to 2) and extremely high risk (2.5 to 9) (Table VII). The observed incidence of mortality among low-risk subjects (−1.5 to −1 point) was 0.28% (2 out of 718 patients), the incidence among moderate-risk subjects was 0.73% (22 out of 3,003 patients) (−0.5 to 0.5 points), the incidence among high-risk participants was 1.17% (22 out of 1,879 patients) (1 to 2 points) and the incidence among extremely high-risk subjects was 6.90% (139 out of 2,015 patients) (Table VII).
Validation stage of the risk score
External validation of the risk score was conducted on a cohort of 10,842 participants (those individuals who underwent craniotomies in 2014–2015). In the validation cohort, the resulting 30-day mortality scores were also divided into four groups according to the quartile of the total risk score as follows: Low risk (−1.5 to −1), moderate risk (−0.5–0.5), high risk (1–2) and extremely high risk (2.5–9). The observed incidence of postoperative 30-day mortality among low-risk participants (−1.5 to −1 point) was 0.29% (3 out of 1,051 patients), among moderate-risk participants it was 1% (41 out of 4,112 patients) (−0.5 to 0.5 points), among high-risk participants it was 2.72% (71 out of 2,609 patients) (1 to 2 points) and among extremely high-risk participants it was 5.64% (158 out of 2,797 patients) (Table VII). The incidences of death in the validation group and the modeling group were similar for each score group (Table VII), thus indicating that the scoring model had good predictive performance.
It was also calculated that the AUC values of the scoring scale model were 0.7844 (95% CI=0.7526–0.8162) and 0.7289 (95% CI=0.7012–0.7566) in the training cohort and the validation cohort, respectively (Table SV). At the best thresholds (2.25 and 1.75), the specificities were 73.54 and 65.49%, and the sensitivities were 75.14 and 68.50% for the training and validation cohorts, respectively (Table SV). The training and validation cohorts both had relatively high NPVs.
Discussion
In the present retrospective cross-sectional study, a personalized prediction nomogram and risk score for postoperative 30-day mortality were developed and validated by evaluating cost-effective and readily available parameters among adult American patients following tumor craniotomy, thus helping clinicians identify individuals at high risk of postoperative 30-day mortality. The prediction model included six parameters: The preoperative WBC count, HCT level, BUN level, age range, functional health status and presence of disseminated cancer. Model evaluation and external validation showed that the nomogram and risk scoring system developed in the present study had excellent predictive performance.
Although numerous death risk prediction models for brain tumors based on demographic, anthropological and clinical information have been established and reported, they have focused mainly on a certain type of brain tumor. A multigene signature has been reported for predicting the prognosis of patients with gliomas (35–39). However, these studies require surgery to obtain pathological tissues from patients to detect genetic signatures. The nomogram developed in the present study differs from those used in these studies in that it is not required to apply genetic signatures derived from tissue analysis for prediction. In addition, Missios et al (40) developed predictive models for postoperative complications (including death) in patients with gliomas on the basis of logistic regression analysis and validated them in a bootstrapped sample. Jia et al (41) performed Cox proportional hazards regression analysis to develop a nomogram to predict the prognosis of meningiomas (World Health Organization Grade III) based on sex, ethnicity, age at diagnosis, histology, tumor site, tumor size, laterality and surgical method. Similarly, previous studies have suggested that the prognostic nomogram comprises factors (age, tumor size and surgery) for overall survival in patients with atypical meningiomas (42). Based on the abovementioned meningioma studies and the present findings, advanced age is indeed a significant risk factor for craniotomy. In terms of brain metastases, prognostic nomograms have been established for breast cancer (43), lung cancer (44,45), bladder cancer (46) and colorectal cancer (47) with brain metastases. All of the abovementioned studies of prediction models were limited to a single type of brain tumor. The present study involved 18,642 patients who underwent craniotomy for a variety of brain tumors and the findings from the training cohort were confirmed in the validation cohort. The AUC values of the nomogram and the scoring model were 0.7949 (95% CI=0.764–0.8255) and 0.7844 (95% CI=0.7526–0.8162), respectively, in the training dataset. Therefore, the clinical applicability of the nomogram and scoring model is broader compared with the relevant studies mentioned above.
A total of 6 risk predictors were identified in the present study, namely the preoperative age range, WBC count, HCT level, BUN level, functional health status and presence of disseminated cancer, for predicting postoperative 30-day mortality in adults with craniotomy for brain tumor. In general, the risk of surgical mortality is increased in older patients. Senders et al (1) suggested that older age and dependent functional status were predictors of postoperative 30-day mortality after craniotomy for primary malignant brain tumors, which is similar to the present findings. Numerous studies have demonstrated that preoperatively lower HCT levels are associated with an increased risk of death after surgery (48–51). Multivariate analysis also demonstrated that preoperative HCT (OR=0.959) was associated with postoperative 30-day mortality (P<0.05), which suggested that a slightly higher HCT level may be a protective factor against 30-day mortality after craniotomy for brain tumor. It was speculated that patients with higher preoperative HCT levels may tolerate a certain degree of blood loss during surgery. Furthermore, elevated BUN levels associated with renal dysfunction are associated with an increased risk of incident diabetes and mortality in patients with cardiovascular disease. A BUN concentration >40 mg/dl was associated with increased mortality in patients who underwent emergency colectomies for Clostridium difficile colitis (52). In the study by Chung et al (53), 6 independent risk factors (including age and preoperative BUN) that are predictive of postoperative 30-day mortality were identified for coronary artery bypass grafts based on the ACS NSQIP database (2005–2010). The present study showed that a BUN level >18.98 (mg/dl) was a risk predictor for postoperative 30-day mortality among adults who underwent craniotomy for brain tumors. Brain metastases are an important cause of mortality and morbidity in patients with cancer (54). This scenario may explain the finding in the present study that disseminated cancer is a risk factor for 30-day mortality after craniotomy. Therefore, the application of the 6 risk predictors in our prediction models was well founded. In addition, the first letter was selected for each risk predictor (excluding HCT; the letter ‘C’ was selected) to name this system as ‘WBC-FAD’ for clinical use.
The present study has several strengths. i) It had a large sample size and the participants originated from multiple centers. ii) A total of 4 prediction models were used, including the LASSO, full, stepwise and MFP models. A simple stepwise model based on the LASSO model was employed. iii) A nomogram and a risk score were simultaneously constructed to ensure model precision and clinical practicability. iv) A formula to calculate the risk of postoperative 30-day mortality was developed based on risk predictors, which can help clinicians quickly and accurately calculate an individual's risk of postoperative 30-day mortality and provide external verification information. v) A complete evaluation of the model was performed for discrimination and calibration. vi) External validation was performed to ensure the reliability of the results.
Although the nomogram and risk score performed well, the present study has several potential limitations. First, it was a secondary retrospective study. The raw data did not reveal other risk factors for mortality, such as characteristics of benign or malignant tumors, lifestyle, pharmacological treatments or socioeconomic factors. However, the present study had a large sample size and the participants were from multiple centers. Our nomogram and risk score had excellent prediction performance in the external validation, thus suggesting that the nomogram and risk score based on the existing 6 risk factors have high generalizability. Second, multiple imputations were used to replace missing values. However, this scenario may lead to bias. Therefore, in the future, it may be considered designing our studies or cooperating with other researchers to collect as many variables as possible as well as reduce missing values. Third, in the present study, the ACS NSQIP database was analyzed from 2012 to 2015 and more valuable models may be obtained by using recent data for data analysis. Fourth, although the performance of the proposed method was tested, real clinical or other related studies are needed before it is widely accepted or applied.
In conclusion, in the present study, a personalized nomogram and risk scoring system (WBC-FAD score) were developed and validated, including the preoperative WBC count, BUN level, HCT level, age range, functional health status and disseminated cancer status, for predicting postoperative 30-day mortality in adults who undergo brain tumor craniotomies in the US. The nomogram and risk score had excellent predictive performance in both the training and validation cohorts for estimating the risk of postoperative 30-day mortality, and they had high generalizability. The categorization of the overall risk relative to the risk status helps to inform the development of mortality of tumor craniotomy intervention or prevention programs. Further improvements in the risk prediction model for tumor craniotomy should consider the nature of the tumor and pharmacological treatments. In future studies these data (including detailed tumor type and tumor location) will be collected to perform stratified analyses and validate our model. Additional clinical and other related studies are needed before this risk scoring system and nomogram for tumor craniotomy can be widely accepted and used.
Supplementary Material
Supporting Data
Supporting Data
Acknowledgements
Not applicable.
Funding
This work was supported by the Shenzhen Second People's Hospital Clinical Research Fund of the Guangdong Province High-level Hospital Construction Project (grant no. 20233357023).
Availability of data and materials
The raw data were obtained from Zhang et al (24) and/or may be downloaded from https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0235273 or from the ACS NSQIP database (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7498000/, S1 Data).
Authors' contributions
YL, HH and GH contributed to the study design and drafted the manuscript. YL, HH, YH and JY were responsible for the statistical analysis. XZ, LC, FC, WL and ZL were also responsible for the statistical analysis, research and interpretation of the data, and revised the manuscript critically. YL, HH and GH confirm the authenticity of all the raw data. All authors have read and approved the final manuscript.
Ethics approval and consent to participate
Patient data were anonymous and previously collected data were analyzed; thus, informed consent was not necessary. Our research was exempted from the Clinical Research Ethics Committee of Shenzhen Second People's Hospital due to the nature of the database (Shenzhen, China; no. 20220407005).
Patient consent for publication
Not applicable.
Competing interests
The authors declare that they have no competing interests.
References
|
Senders JT, Muskens IS, Cote DJ, Goldhaber NH, Dawood HY, Gormley WB, Broekman MLD and Smith TR: Thirty-Day outcomes after craniotomy for primary malignant brain tumors: A national surgical quality improvement program analysis. Neurosurgery. 83:1249–1259. 2018. View Article : Google Scholar : PubMed/NCBI | |
|
De la Garza-Ramos R, Kerezoudis P, Tamargo RJ, Brem H, Huang J and Bydon M: Surgical complications following malignant brain tumor surgery: An analysis of 2002–2011 data. Clin Neurol Neurosurg. 140:6–10. 2016. View Article : Google Scholar : PubMed/NCBI | |
|
Lonjaret L, Guyonnet M, Berard E, Vironneau M, Peres F, Sacrista S, Ferrier A, Ramonda V, Vuillaume C, Roux FE, et al: Postoperative complications after craniotomy for brain tumor surgery. Anaesth Crit Care Pain Med. 36:213–218. 2017. View Article : Google Scholar : PubMed/NCBI | |
|
Writing Committee for the VISION Study Investigators, . Devereaux PJ, Biccard BM, Sigamani A, Xavier D, Chan MTV, Srinathan SK, Walsh M, Abraham V, Pearse R, et al: Association of postoperative High-Sensitivity troponin levels with myocardial injury and 30-Day mortality among patients undergoing noncardiac surgery. JAMA. 317:1642–1651. 2017. View Article : Google Scholar : PubMed/NCBI | |
|
Fritz BA, Cui Z, Zhang M, He Y, Chen Y, Kronzer A, Ben Abdallah A, King CR and Avidan MS: Deep-learning model for predicting 30-day postoperative mortality. Br J Anaesth. 123:688–695. 2019. View Article : Google Scholar : PubMed/NCBI | |
|
Watters DA, Hollands MJ, Gruen RL, Maoate K, Perndt H, McDougall RJ, Morriss WW, Tangi V, Casey KM and McQueen KA: Perioperative mortality rate (POMR): A global indicator of access to safe surgery and anaesthesia. World J Surg. 39:856–864. 2015. View Article : Google Scholar : PubMed/NCBI | |
|
Lochte BC, Carroll KT, Hirshman B, Lanman T, Carter B and Chen CC: Smoking as a risk factor for postcraniotomy 30-Day mortality. World Neurosurg. 127:e400–e406. 2019. View Article : Google Scholar : PubMed/NCBI | |
|
Williams M, Treasure P, Greenberg D, Brodbelt A and Collins P: Surgeon volume and 30 day mortality for brain tumours in England. Br J Cancer. 115:1379–1382. 2016. View Article : Google Scholar : PubMed/NCBI | |
|
Dikmen S, Machamer J, Manley GT, Yuh EL, Nelson LD and Temkin NR; TRACK-TBI Investigators, : Functional status examination versus glasgow outcome scale extended as outcome measures in traumatic brain injuries: How do they compare? J Neurotrauma. 36:2423–2429. 2019. View Article : Google Scholar : PubMed/NCBI | |
|
Ois A, Vivas E, Figueras-Aguirre G, Guimaraens L, Cuadrado-Godia E, Avellaneda C, Bertran-Recasens B, Rodríguez-Campello A, Gracia MP, Villalba G, et al: Misdiagnosis worsens prognosis in subarachnoid hemorrhage with good hunt and hess score. Stroke. 50:3072–3076. 2019. View Article : Google Scholar : PubMed/NCBI | |
|
Khalil H, Aldaajani ZF, Aldughmi M, Al-Sharman A, Mohammad T, Mehanna R, El-Jaafary SI, Dahshan A, Ben Djebara M, Kamel WA, et al: Validation of the arabic version of the movement disorder Society-Unified parkinson's disease rating scale. Mov Disord. 37:826–841. 2022. View Article : Google Scholar : PubMed/NCBI | |
|
Gittleman H, Lim D, Kattan MW, Chakravarti A, Gilbert MR, Lassman AB, Lo SS, Machtay M, Sloan AE, Sulman EP, et al: An independently validated nomogram for individualized estimation of survival among patients with newly diagnosed glioblastoma: NRG Oncology RTOG 0525 and 0825. Neuro Oncol. 19:669–677. 2017.PubMed/NCBI | |
|
Mijderwijk HJ, Nieboer D, Incekara F, Berger K, Steyerberg EW, van den Bent MJ, Reifenberger G, Hänggi D, Smits M, Senft C, et al: Development and external validation of a clinical prediction model for survival in patients with IDH wild-type glioblastoma. J Neurosurg. Jan 14–2022.(Epub ahead of print). View Article : Google Scholar : PubMed/NCBI | |
|
Wang Z, Gao L, Guo X, Feng C, Lian W, Deng K and Xing B: Development of a nomogram with alternative splicing signatures for predicting the prognosis of glioblastoma: A study based on Large-Scale sequencing data. Front Oncol. 10:12572020. View Article : Google Scholar : PubMed/NCBI | |
|
Molinaro AM, Wrensch MR, Jenkins RB and Eckel-Passow JE: Statistical considerations on prognostic models for glioma. Neuro Oncol. 18:609–623. 2016. View Article : Google Scholar : PubMed/NCBI | |
|
Li N, Mo Y, Huang C, Han K, He M, Wang X, Wen J, Yang S, Wu H, Dong F, et al: A clinical semantic and radiomics nomogram for predicting brain invasion in WHO grade II meningioma based on tumor and Tumor-to-Brain interface features. Front Oncol. 11:7521582021. View Article : Google Scholar : PubMed/NCBI | |
|
Zhang J, Yao K, Liu P, Liu Z, Han T, Zhao Z, Cao Y, Zhang G, Zhang J, Tian J and Zhou J: A radiomics model for preoperative prediction of brain invasion in meningioma non-invasively based on MRI: A multicentre study. Ebiomedicine. 58:1029332020. View Article : Google Scholar : PubMed/NCBI | |
|
Pietrantonio F, Aprile G, Rimassa L, Franco P, Lonardi S, Cremolini C, Biondani P, Sbicego EL, Pasqualetti F, Tomasello G, et al: A new nomogram for estimating survival in patients with brain metastases secondary to colorectal cancer. Radiother Oncol. 117:315–321. 2015. View Article : Google Scholar : PubMed/NCBI | |
|
Marko NF, Xu Z, Gao T, Kattan MW and Weil RJ: Predicting survival in women with breast cancer and brain metastasis: A nomogram outperforms current survival prediction models. Cancer. 118:3749–3757. 2012. View Article : Google Scholar : PubMed/NCBI | |
|
Cheng S, Yang L, Dai X, Wang J and Han X: The risk and prognostic factors for brain metastases in esophageal cancer patients: An analysis of the SEER database. BMC Cancer. 21:10572021. View Article : Google Scholar : PubMed/NCBI | |
|
Zhai Y, Bai J, Li M, Wang S, Li C, Wei X and Zhang Y: A nomogram to predict the progression-free survival of clival chordoma. J Neurosurg. 134:144–152. 2019. View Article : Google Scholar : PubMed/NCBI | |
|
Dasgupta A, Gupta T, Pungavkar S, Shirsat N, Epari S, Chinnaswamy G, Mahajan A, Janu A, Moiyadi A, Kannan S, et al: Nomograms based on preoperative multiparametric magnetic resonance imaging for prediction of molecular subgrouping in medulloblastoma: Results from a radiogenomics study of 111 patients. Neuro Oncol. 21:115–124. 2019. View Article : Google Scholar : PubMed/NCBI | |
|
Zhang D, Zhuo H, Yang G, Huang H, Li C, Wang X, Zhao S, Moliterno J and Zhang Y: Postoperative pneumonia after craniotomy: Incidence, risk factors and prediction with a nomogram. J Hosp Infect. 105:167–175. 2020. View Article : Google Scholar : PubMed/NCBI | |
|
Zhang J, Li YI, Pieters TA, Towner J, Li KZ, Al-Dhahir MA, Childers F and Li YM: Sepsis and septic shock after craniotomy: Predicting a significant patient safety and quality outcome measure. PLoS One. 15:e2352732020. | |
|
Groenwold RH, White IR, Donders AR, Carpenter JR, Altman DG and Moons KG: Missing covariate data in clinical research: When and when not to use the missing-indicator method for analysis. CMAJ. 184:1265–1269. 2012. View Article : Google Scholar : PubMed/NCBI | |
|
White IR, Royston P and Wood AM: Multiple imputation using chained equations: Issues and guidance for practice. Stat Med. 30:377–399. 2011. View Article : Google Scholar : PubMed/NCBI | |
|
Friedman J, Hastie T and Tibshirani R: Regularization paths for generalized linear models via coordinate descent. J Stat Softw. 33:1–22. 2010. View Article : Google Scholar : PubMed/NCBI | |
|
Kidd AC, McGettrick M, Tsim S, Halligan DL, Bylesjo M and Blyth KG: Survival prediction in mesothelioma using a scalable Lasso regression model: Instructions for use and initial performance using clinical predictors. BMJ Open Respir Res. 5:e0002402018. View Article : Google Scholar : PubMed/NCBI | |
|
Della Rosa PA, Miglioli C, Caglioni M, Tiberio F, Mosser KHH, Vignotto E, Canini M, Baldoli C, Falini A, Candiani M and Cavoretto P: A hierarchical procedure to select intrauterine and extrauterine factors for methodological validation of preterm birth risk estimation. BMC Pregnancy Childbirth. 21:3062021. View Article : Google Scholar : PubMed/NCBI | |
|
Roh J, Jung J, Lee Y, Kim SW, Pak HK, Lee AN, Lee J, Cho J, Cho H, Yoon DH, et al: Risk stratification using multivariable fractional polynomials in diffuse large B-Cell lymphoma. Front Oncol. 10:3292020. View Article : Google Scholar : PubMed/NCBI | |
|
Weng ZA, Huang XX, Deng D, Yang ZG, Li SY, Zang JK, Li YF, Liu YF, Wu YS, Zhang TY, et al: A new nomogram for predicting the risk of intracranial hemorrhage in acute ischemic stroke patients after intravenous thrombolysis. Front Neurol. 13:7746542022. View Article : Google Scholar : PubMed/NCBI | |
|
Alba AC, Agoritsas T, Walsh M, Hanna S, Iorio A, Devereaux PJ, McGinn T and Guyatt G: Discrimination and calibration of clinical prediction models: Users' guides to the medical literature. JAMA. 318:1377–1384. 2017. View Article : Google Scholar : PubMed/NCBI | |
|
Mehta HB, Mehta V, Girman CJ, Adhikari D and Johnson ML: Regression coefficient-based scoring system should be used to assign weights to the risk index. J Clin Epidemiol. 79:22–28. 2016. View Article : Google Scholar : PubMed/NCBI | |
|
Collins GS, Reitsma JB, Altman DG and Moons KG: Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): The TRIPOD statement. BMJ. 350:g75942015. View Article : Google Scholar : PubMed/NCBI | |
|
Hu X, Martinez-Ledesma E, Zheng S, Kim H, Barthel F, Jiang T, Hess KR and Verhaak RGW: Multigene signature for predicting prognosis of patients with 1p19q co-deletion diffuse glioma. Neuro Oncol. 19:786–795. 2017. View Article : Google Scholar : PubMed/NCBI | |
|
Zhang Y, Ma W, Fan W, Ren C, Xu J, Zeng F, Bao Z, Jiang T and Zhao Z: Comprehensive transcriptomic characterization reveals core genes and module associated with immunological changes via 1619 samples of brain glioma. Cell Death Dis. 12:11402021. View Article : Google Scholar : PubMed/NCBI | |
|
Zheng Y, Ji Q, Xie L, Wang C, Yu CN, Wang YL, Jiang J, Chen F and Li WB: Ferroptosis-related gene signature as a prognostic marker for lower-grade gliomas. J Cell Mol Med. 25:3080–3090. 2021. View Article : Google Scholar : PubMed/NCBI | |
|
Wang X, Gao M, Ye J, Jiang Q, Yang Q, Zhang C, Wang S, Zhang J, Wang L, Wu J, et al: An immune Gene-Related Five-lncRNA signature for to predict glioma prognosis. Front Genet. 11:6120372020. View Article : Google Scholar : PubMed/NCBI | |
|
Yun D, Wang X, Wang W, Ren X, Li J, Wang X, Liang J, Liu J, Fan J, Ren X, et al: A novel prognostic signature based on glioma essential Ferroptosis-Related genes predicts clinical outcomes and indicates treatment in glioma. Front Oncol. 12:8977022022. View Article : Google Scholar : PubMed/NCBI | |
|
Missios S, Kalakoti P, Nanda A and Bekelis K: Craniotomy for glioma resection: A predictive model. World Neurosurg. 83:957–964. 2015. View Article : Google Scholar : PubMed/NCBI | |
|
Jia Z, Yan Y, Wang J, Yang H, Zhan H, Chen Q, He Y and Hu Y: Development and validation of prognostic nomogram in patients with WHO grade III meningioma: A retrospective cohort study based on SEER database. Front Oncol. 11:7199742021. View Article : Google Scholar : PubMed/NCBI | |
|
Zhang GJ, Liu XY and You C: Clinical factors and outcomes of atypical meningioma: A Population-Based study. Front Oncol. 11:6766832021. View Article : Google Scholar : PubMed/NCBI | |
|
Xiong Y, Cao H, Zhang Y, Pan Z, Dong S, Wang G, Wang F and Li X: Nomogram-Predicted survival of breast cancer brain metastasis: A SEER-Based population study. World Neurosurg. 128:e823–e834. 2019. View Article : Google Scholar : PubMed/NCBI | |
|
Zindler JD, Jochems A, Lagerwaard FJ, Beumer R, Troost EGC, Eekers DBP, Compter I, van der Toorn PP, Essers M, Oei B, et al: Individualized early death and long-term survival prediction after stereotactic radiosurgery for brain metastases of non-small cell lung cancer: Two externally validated nomograms. Radiother Oncol. 123:189–194. 2017. View Article : Google Scholar : PubMed/NCBI | |
|
Shen H, Deng G, Chen Q and Qian J: The incidence, risk factors and predictive nomograms for early death of lung cancer with synchronous brain metastasis: A retrospective study in the SEER database. BMC Cancer. 21:8252021. View Article : Google Scholar : PubMed/NCBI | |
|
Yao Z, Zheng Z, Ke W, Wang R, Mu X, Sun F, Wang X, Garg S, Shi W, He Y and Liu Z: Prognostic nomogram for bladder cancer with brain metastases: A National Cancer Database analysis. J Transl Med. 17:4112019. View Article : Google Scholar : PubMed/NCBI | |
|
Nieder C, Hintz M and Grosu AL: Predicted survival in patients with brain metastases from colorectal cancer: Is a current nomogram helpful? Clin Neurol Neurosurg. 143:107–110. 2016. View Article : Google Scholar : PubMed/NCBI | |
|
Bodewes T, Pothof AB, Darling JD, Deery SE, Jones DW, Soden PA, Moll FL and Schermerhorn ML: Preoperative anemia associated with adverse outcomes after infrainguinal bypass surgery in patients with chronic limb-threatening ischemia. J Vasc Surg. 66:1775–1785.e2. 2017. View Article : Google Scholar : PubMed/NCBI | |
|
Kouyoumdjian A, Trepanier M, Al Shehhi R, Cools-Lartigue J, Ferri LE, Lee L and Mueller CL: The effect of preoperative anemia and perioperative transfusion on surgical outcomes after gastrectomy for gastric cancer. J Surg Res. 259:523–531. 2021. View Article : Google Scholar : PubMed/NCBI | |
|
Faraoni D, DiNardo JA and Goobie SM: Relationship between preoperative anemia and In-Hospital mortality in children undergoing noncardiac surgery. Anesth Analg. 123:1582–1587. 2016. View Article : Google Scholar : PubMed/NCBI | |
|
Zhang X, Zhang F, Qiao W, Zhang X, Zhao Z and Li M: Low hematocrit is a strong predictor of poor prognosis in lung cancer patients. Biomed Res Int. 2018:68049382018.PubMed/NCBI | |
|
Lee DY, Chung EL, Guend H, Whelan RL, Wedderburn RV and Rose KM: Predictors of mortality after emergency colectomy for Clostridium difficile colitis: An analysis of ACS-NSQIP. Ann Surg. 259:148–156. 2014. View Article : Google Scholar : PubMed/NCBI | |
|
Chung PJ, Carter TI, Burack JH, Tam S, Alfonso A and Sugiyama G: Predicting the risk of death following coronary artery bypass graft made simple: A retrospective study using the American College of Surgeons National Surgical quality improvement program database. J Cardiothorac Surg. 10:622015. View Article : Google Scholar : PubMed/NCBI | |
|
Cagney DN, Martin AM, Catalano PJ, Redig AJ, Lin NU, Lee EQ, Wen PY, Dunn IF, Bi WL, Weiss SE, et al: Incidence and prognosis of patients with brain metastases at diagnosis of systemic malignancy: A population-based study. Neuro Oncol. 19:1511–1521. 2017. View Article : Google Scholar : PubMed/NCBI |
