Prediction method based on hypospadias individualized operative selection prediction model

Abstract

The invention provides a prediction method based on a hypospadias individualized surgical selection prediction model, belongs to the technical field of model prediction, and aims to construct a hypospadias individualized surgical selection strategy flow chart which organically integrates hypospadias penis anatomy data into a surgical selection strategy, so that the prediction accuracy is improved. Especially, a prediction model established for judging whether the urethral plate is transected or not in an operation has good stability through external verification, and various data of the prediction model are simple, objective and easy to obtain. Datamation selection is provided for operation selection after the urethral plate is reserved and the urethral plate is transected, and the flow chart can provide objective and quantifiable operation type selection reference basis for a beginner of a hypospadias operation.

Description

Technical Field

[0001] The present invention relates to the field of model prediction technology, and in particular to a prediction method based on an individualized hypospadias surgical procedure selection prediction model. Background Art

[0002] Hypospadias is a common congenital disorder of the male genitourinary system, with an incidence of approximately 0.3% to 0.9%, and a recent trend of increasing incidence. Surgery is the only treatment for hypospadias. However, unlike other organ surgeries, hypospadias surgery must balance both function and aesthetics. Therefore, hypospadias surgery requires both precise technical mastery and an art form requiring in-depth exploration. Over 300 urethroplasty procedures have been published, but no single procedure can repair all types of hypospadias.

[0003] Although various surgical procedures are available for hypospadias, overall treatment efficacy remains to be improved, particularly for proximal hypospadias with severe penile chordee. Foreign literature reports a 5%-10% postoperative complication rate for distal hypospadias, while the complication rate for proximal hypospadias remains high, ranging from 15% to 90%, regardless of surgical approach. Key factors influencing the prognosis of hypospadias are related to the patient's penile anatomy and the surgical approach used.

[0004] Studies have shown that the learning curve for surgical repair of proximal hypospadias stabilizes after at least 50 cases. Beginners in hypospadias surgery often seek a single procedure to address all patients, ignoring individual differences. Surgical strategy often determines the success or failure of the procedure. Therefore, for beginners in hypospadias surgery, in addition to gradually mastering various urethroplasty techniques and developing proficient surgical skills, it is crucial to develop a strategy for individualizing surgical plans based on the patient's penile anatomy to truly improve the treatment outcome of hypospadias.

[0005] The key components of hypospadias surgery include correction of chordee and urethroplasty. In 2022, Castagnetti et al. proposed a modified "stepwise" approach to correct chordee, which can serve as a reference for surgeons seeking intraoperative correction of chordee. Regarding the surgical selection strategy for urethroplasty, some have proposed that for distal hypospadias, tubularized incised plate urethroplasty (TIP) and Mathieu urethroplasty with a urethral meatus base flap have similar postoperative complication rates and urine flow rates, but TIP has a more aesthetically pleasing appearance. For proximal hypospadias, when chordal flexion is <30°, current data show that TIP and onlay island flap urethroplasty have similar complication rates. However, when chordal flexion is >30°, urethral plate transection is required to perform transverse preputial island flap urethroplasty (TPIF), pedicled preputial flap urethroplasty with the urethral meatus as the base (Koyanagi), or staged Byars flap. Similarly, the consensus among domestic hypospadias experts also proposes a preliminary surgical procedure selection strategy based on whether the urethral plate is retained during surgery. However, there is currently no comprehensive and reasonable "data-based" flow chart for urethroplasty procedure selection strategy in China or abroad.

[0006] Therefore, this field urgently needs a technical solution that can establish a data-based individualized surgical selection strategy for hypospadias suitable for domestic children.

[0007] The information disclosed in this background technology section is only intended to enhance understanding of the overall background of the invention and should not be considered as an admission or any form of suggestion that the information constitutes the prior art already known to a person skilled in the art. Summary of the Invention

[0008] The purpose of the present invention is to provide a method for establishing a data-based individualized surgical selection strategy for hypospadias suitable for domestic children.

[0009] To achieve the above object, the present invention provides the following solutions:

[0010] A prediction method based on a prediction model for individualized surgical procedure selection for hypospadias, including:

[0011] Step 1: Collect penile anatomical data;

[0012] Step 2: Combined with the preoperative penile anatomy data, a clinical prediction model for predicting intraoperative urethral plate transection was established;

[0013] Step 3: Based on the intraoperative penile anatomical data, determine the most appropriate penile anatomy for various surgical procedures in patients with preserved urethral plates, and determine the most appropriate penile anatomy for single-stage and staged procedures in patients with transected urethral plates.

[0014] Step 4: Integrate steps 2 and 3 to establish a data-based flowchart for individualized surgical selection strategies for children with hypospadias;

[0015] Step 5: Use statistical software to organize and analyze data, draw a nomogram of the clinical prediction model, perform internal validation, explain the discrimination of the model, test the calibration of the model and draw a calibration curve, evaluate the consistency between the model's predicted risk and the actual risk, and use the clinical decision curve to evaluate the clinical benefit of the model. The height reflects the size of the net benefit, and the width reflects the breadth of the model's scope of application.

[0016] Optionally, the inclusion criteria for the penis anatomy data include:

[0017] Children with hypospadias who are undergoing first diagnosis and treatment, have complete penile anatomical data, and have complete follow-up information;

[0018] Exclusion criteria were: hypospadias with intact foreskin, previous history of circumcision or urethra-related surgery, missing key penile anatomical data, chromosomal or genetic abnormalities, combined sexual developmental abnormalities, or incomplete follow-up data.

[0019] Optionally, the clinical prediction model is constructed using a hospital data cohort as a training set, and a multicenter cohort as a validation set for external validation of the clinical prediction model.

[0020] Optionally, the surgical procedures include: MAGPI, TIP, and Onlay procedures.

[0021] Optionally, the step 5 is specifically as follows:

[0022] The data is arranged and analyzed by SPSS 23.0 statistical software, the nomogram function of the rms package in the R Studio statistical software is used to draw the nomogram of the clinical prediction model; first, the normality test is performed on the continuous variables, and the test results are that the age, the length of the penis, the length of the glans penis, the width of the glans penis, the width of the urethral plate, the width of the urethral plate after longitudinal cutting, the length of the urethral defect, the number of the lower curvature of the penis and the UDR value all do not meet the normal distribution, then the frequency is used for variable description, and the chi-square test or Fisher's exact probability method is used for group comparison, the continuous variables that do not meet the normal distribution are represented by the median, and whether there is a statistical difference between the two groups is compared by using wilcox test; whether there is a complication after hypospadias repair is used as the dependent variable, and non-parametric test is used for group comparison; the continuous variables are analyzed by using Spearman correlation, when r<0.4, it is considered that the correlation is weak, when r is between 0.4-0.7, it is considered that the correlation is moderate, and when r>0.7, it is considered that the correlation is high;

[0023] The influencing factors with P<0.05 in the single factor Logistic regression analysis are included in the multi-factor Logistic regression analysis, the multi-factor analysis is used for accurate estimation of the influence of multiple influencing factors, and the influence of confounding factors on the analysis result is avoided; the Logistic regression method is used for analyzing the risk factors of postoperative complications of hypospadias, the odds ratio and the 95% confidence interval are calculated; the forward stepwise method is used to select the independent variables finally introduced into the prediction model; the variance inflation coefficient is used to evaluate whether the prediction factors included in the model have multicollinearity, the VIF value is greater than 1, and the closer to 1, the lighter the multicollinearity, and vice versa, and the VIF less than 10 is acceptable; after the nomogram of the clinical prediction model is established, in order to further evaluate the effect of the nomogram model, 1000 times of Bootstrap self-sampling method is used for internal verification, the ROC curve is drawn by using the pROC package, and the AUC under the ROC is used to explain the discrimination of the model, the AUC value ranges from 0 to 1, the closer to 1, the better the model discrimination ability, the AUC equal to 0.5 indicates that the model has no prediction ability, less than 0.5 indicates that the model prediction is opposite to the actual result, and AUC<0.6 is considered to have poor discrimination, AUC value is 0.6-0.75, it is considered that the model has certain discrimination ability, and AUC>0.75 is considered to have good discrimination; the calibration degree of the prediction model is evaluated by using the Hosmer-Lemeshow goodness-of-fit test, and the calibration curve is drawn, the consistency between the model calibration line and the standard line is closer, and the calibration degree is better; the clinical benefit of the model is evaluated by using the clinical decision curve, which highly reflects the size of the net benefit and the width of the model application range.

[0024] Compared with the prior art, the present application has the following advantages:

[0025] The present invention provides a prediction method based on a personalized hypospadias surgical procedure selection prediction model, enabling the construction of a flowchart for personalized hypospadias surgical selection strategies. This flowchart organically integrates hypospadias penile anatomical data into the surgical procedure selection strategy. In particular, the prediction model established for determining whether to transect the urethral plate during surgery has been externally validated to demonstrate good stability, and the data in the prediction model are simple, objective, and easily accessible. Furthermore, it provides a digitized choice of surgical options for retaining the urethral plate versus transecting the urethral plate. This flowchart can provide an objective and quantifiable reference for surgical procedure selection for beginners in hypospadias surgery. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0027] Figure 1 A nomogram for transecting the urethral plate during hypospadias surgery provided in an embodiment of the present invention.

[0028] Figure 2 Schematic diagram of the calibration curve of the clinical prediction model for transecting the urethral plate during hypospadias surgery provided by an embodiment of the present invention.

[0029] Figure 3 Schematic diagram of the ROC curve of the clinical prediction model training set for intraoperative urethral plate transecting provided in an embodiment of the present invention.

[0030] Figure 4 This is the ROC curve of the validation set of the clinical prediction model for intraoperative urethral plate transection provided by an embodiment of the present invention.

[0031] Figure 5 Schematic diagram of the DCA curve of the clinical prediction model training set of the transverse urethral plate provided in an embodiment of the present invention.

[0032] Figure 6 A comparison chart of the number of surgeries for different surgical groups in children with urethral defects greater than 3.55 cm in length provided by an embodiment of the present invention.

[0033] Figure 7 A flow chart of a personalized surgical selection strategy for hypospadias based on multi-center penile anatomical characteristics provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0034] With reference to the accompanying drawings, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those ordinarily skilled in the art without creative effort belong to the scope of the present application.

[0035] The present application aims to provide a method capable of establishing a data-based individualized surgical selection strategy for children with hypospadias.

[0036] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0037] Embodiment 1

[0038] The present embodiment provides a prediction method based on a hypospadias individualized surgical selection prediction model, comprising:

[0039] Step 1: collecting penile anatomy data;

[0040] Step 2: combining the preoperative penile anatomy data, establishing a clinical prediction model for predicting the intraoperative transverse urethral plate;

[0041] Step 3: combining the intraoperative penile anatomy data, in children with preserved urethral plate, determining the most suitable penile anatomy condition corresponding to various surgical procedures, and in children with transverse urethral plate, determining the most suitable penile anatomy condition corresponding to one-stage surgical procedures and staged surgical procedures;

[0042] Step 4: integrating Step 2 and Step 3, establishing a data-based individualized surgical selection strategy flowchart for children with hypospadias;

[0043] Step 5: using statistical software to arrange and analyze the data, drawing a nomogram of the clinical prediction model, performing internal validation, explaining the discrimination of the model, testing the calibration of the model and drawing a calibration curve, evaluating the consistency of the predicted risk and the actual risk of the model, using a clinical decision curve to evaluate the clinical benefit of the model, and highly reflecting the size of net benefit and the width of the applicable range of the model.

[0044] In one embodiment, the inclusion criteria of the penile anatomy data include:

[0045] Children with hypospadias who are first diagnosed and treated, have complete penile anatomy data, and have complete follow-up data;

[0046] Exclusion criteria were: hypospadias with intact foreskin, previous history of circumcision or urethra-related surgery, missing key penile anatomical data, chromosomal or genetic abnormalities, combined sexual developmental abnormalities, or incomplete follow-up data.

[0047] In one embodiment, the clinical prediction model is constructed using a hospital data cohort as a training set, and a multi-center cohort is used as a validation set to perform external validation of the clinical prediction model.

[0048] In one embodiment, the procedures include: MAGPI, TIP, and Onlay procedures.

[0049] In one embodiment, step 5 is specifically as follows: SPSS 23.0 statistical software is used for data collation and analysis, and the nomogram function of the "rms" package in R Studio statistical software is used to draw a nomogram of the clinical prediction model. A normality test (Shapiro-Wilk test) is first performed on the continuous variables. It is found that the continuous variables such as age, penis length, glans penis length, glans penis width, urethral plate width, urethral plate width after longitudinal incision, urethral defect length, penis curvature number, and UDR value do not meet the normal distribution. Therefore, in this embodiment, categorical variables were described as frequencies (percentages), and intergroup comparisons were performed using the Chi-square test or Fisher's exact probability method. Continuous variables that did not conform to a normal distribution were expressed as medians (25% quantiles, 75% quantiles) [M(P25, P75)]. The Wilcox test was used to compare whether there was a statistical difference between the two groups. The presence or absence of complications after hypospadias surgery was used as the dependent variable, and the non-parametric test (Mann-Whitney U test) was used for intergroup comparisons. Continuous variables were analyzed using Spearman correlation analysis. When r < 0.4, it was considered a weak correlation, r between 0.4-0.7 was a moderate correlation, and r > 0.7 was a high correlation. In all cases, a P value < 0.05 was considered statistically significant.

[0050] The factors with P<0.05 in the single-factor Logistic regression analysis were included in the multi-factor Logistic regression analysis. The multi-factor analysis was used to accurately estimate the influence of multiple factors and avoid the influence of confounding factors on the analysis results. The Logistic regression method was used to analyze the risk factors of postoperative complications of hypospadias. The odds ratio (OR) and 95% confidence interval (CI) were calculated, and P<0.05 was considered statistically significant. The forward stepwise method was used to select the independent variables finally introduced into the prediction model. The variance inflation factor (VIF) was used to evaluate whether there was multicollinearity among the prediction factors included in the model. The VIF value greater than 1, the closer to 1, the lighter the multicollinearity, and vice versa. Generally, VIF less than 10 is considered acceptable. After establishing the nomogram of the clinical prediction model, 1000 times of Bootstrap self-sampling method was used for internal validation. The ROC curve was drawn using the "pROC" package, and the AUC under ROC (i.e. C index) was used to explain the discrimination of the model. The AUC value ranges from 0 to 1, the closer to 1, the better the model discrimination ability, and the AUC equal to 0.5 indicates that the model has no prediction ability, and less than 0.5 indicates that the model prediction is opposite to the actual result. Generally, AUC<0.6 is considered to have poor discrimination, AUC value of 0.6-0.75 is considered to have certain discrimination, and AUC>0.75 is considered to have good discrimination. The Hosmer-Lemeshow goodness-of-fit test was used to evaluate the calibration of the prediction model, and the calibration curve (Calibration) was drawn to evaluate the consistency of the model prediction risk and the actual risk. The closer the calibration line of the model to the standard line, the better the calibration. The decision curve analysis (DCA) curve was used to evaluate the clinical benefit of the model, which highly reflects the size of net benefit and the width reflects the width of the model application range.

[0051] Penile anatomical data measurement and surgical approach:

[0052] All children in this example were measured according to the penile anatomical data measurement standard. The specific measurement method is described in detail in the first part of the study materials and methods. The surgical process is as follows:

[0053] The length of the penis (cm), the width of the glans penis (mm), the length of the glans penis (mm), the width of the urethral plate (mm), the preoperative penile curvature (°), and the preoperative penile curvature (°) were measured before the operation, and the hat-shaped prepuce morphology (single eye, double eye, irregular), the flatness of the urethral plate, the elasticity of the urethral plate, the scrotal morphology, the morphology of the urethral opening before the correction of the penile curvature, and the position of the urethral opening before the correction of the penile curvature were recorded.

[0054] Then, a traction line is placed on the glans penis to measure the presence of a membranous urethra and its length (mm). A parallel longitudinal incision is made from the urethral opening to the navicular fossa along both sides of the urethral plate. A circular incision is made on the inner plate of the foreskin 1.0 cm away from the coronal sulcus, and the urethral plate is retained. The penile skin and subcutaneous tissue are degloved and freed to the root of the penis. The degree of penile curvature after simple degloving is measured (°). The ventral fibrous tissue is then released, and an artificial erection test is performed. The angle of penile curvature is measured with a protractor. If penile curvature is still present but less than 30°, the urethral plate is retained, the dorsal vascular and nerve bundle is freed, and the dorsal tunica albuginea is tightened with non-absorbable sutures. Then, an artificial erection test is performed again to determine whether the penile curvature is fully corrected. If the chordee is completely corrected at this time, the urethral defect length (UD) is measured, the urethral defect ratio (UDR) is calculated, and urethroplasty with urethral plate preservation is performed, such as MAGPI, TIP, Onlay, Mathieu, etc.

[0055] If, after degloving the penile skin and releasing the ventral fibrous tissue, an artificial erection test is performed, and a chord of greater than 30° is still present, as measured using a protractor, the surgeon must determine whether to retain the urethral plate. ① If the decision is to retain the urethral plate, dorsal tunica albuginea tightening is performed to further assess the effectiveness of chord correction. If chord is not present, urethroplasty with the urethral plate retained can be performed. If chord is still present, the urethral plate should be transected, supplemented with dorsal tunica albuginea tightening if necessary to completely correct chord. Procedures include TPIF, TPIF with Duplay, Koyanagi, and staged TPIF. ② If the decision is to transect the urethral plate, the presence of chord is assessed after transection. If chord is still present, dorsal tunica albuginea tightening is performed to further correct chord before urethroplasty with the urethral plate transected.

[0056] Follow-up methods:

[0057] Follow-up was conducted through outpatient consultations and telephone interviews to understand whether the children had complications such as urethral fistula, urethral stenosis, urethral diverticulum-like dilatation, and penile curvature after surgery that required reoperation. The number of unplanned reoperations after the first urethroplasty was recorded. For example, if a child who underwent TPIF surgery developed urethral stenosis after surgery and repeated urethral dilatation twice did not improve, and the child underwent urethral fistula repair after urethral incision, the number of unplanned reoperations was recorded as 4; if a child who underwent staged TPIF surgery did not have complications after the second-stage surgery, the number of unplanned reoperations for this child was 0.

[0058] result:

[0059] A total of 1500 children with hypospadias initially treated at 16 tertiary clinical centers met the inclusion criteria (819 in the hospital cohort and 681 in the multicenter cohort). Among them, 69 underwent MAGPI, 493 TIP, 138 Onlay, 15 Mathieu, 543 TPIF, 45 TPIF+Duplay, 57 Koyanagi, and 140 underwent staged procedures (13 with the Byars flap and 127 with staged TPIF).

[0060] A total of 1500 patients were divided into a urethral plate retention group (MAGPI, TIP, Onlay, and Mathieu procedures), totaling 715 patients (47.7%), and a urethral plate transection group (TPIF, TPIF+Duplay, Koyanagi, and staged procedures), totaling 785 patients (52.3%). Baseline clinical data for the two groups are shown in Table 1. Significant differences were found between the two groups in age at surgery, penile length, glans length and width, foreskin morphology, superficial foreskin vascularity, urethral plate width, urethral opening position before and after correction of chordee, Barcat classification, urethral defect length, UDR value, and degree of chordee (P < 0.05). The urethral plate width in the urethral plate retention group was 5 (4, 6) mm, and the penile curvature was 30 (20, 45) degrees. The urethral plate width in the urethral plate transection group was 4 (3, 5) mm, and the penile curvature was 60 (50, 80) degrees. Univariate logistic regression analysis was performed on the clinical data of 1500 children, with urethral plate transection as the dependent variable (see Table 2). Statistically significant indicators (P < 0.05) were selected and included in the multivariate logistic regression analysis. The results showed that urethral plate width, urethral defect length, and penile curvature (degrees) were influencing factors for urethral plate transection (P < 0.05). (See Table 2 for details).

[0061] Table 1 Baseline data of 1500 children in the urethral plate retention group and the urethral plate transection group

[0062]

[0063]

[0064] Note: Continuous variables are expressed as median (25% quantile, 75% quantile) [M(P25, P75)]; categorical variables are expressed as frequency (percentage); UDR, urethral defect ratio.

[0065] Table 2 Analysis of risk factors for urethral plate transection in 1500 children

[0066]

[0067]

[0068] Note: UDR, urethral defect ratio.

[0069] All patients in this example were followed up for 42 months, and a total of 481 patients developed complications, with a complication rate of 32.1% (481 / 1500). These complications included urethral fistula in 333 cases (22.2%), urethral stricture in 156 cases (10.4%), urethral diverticulum in 85 cases (5.7%), and recurrent chordee in 4 cases (0.3%). All patients who developed complications underwent reoperation. Based on whether the urethral plate was retained during surgery, the 1500 patients were divided into a urethral plate retention group (715 patients) and a urethral plate transection group (785 patients). The postoperative complication rate in the urethral plate retention group was 24.76%, while the postoperative complication rate in the urethral plate transection group was 38.73%, with a statistically significant difference between the two groups (χ² = 33.527, P < 0.001).

[0070] The complication rate of MAGPI procedure was 4.3% (3 / 69), the complication rate of TIP procedure was 23.7% (117 / 493), the complication rate of Onlay procedure was 37.7% (52 / 138), the complication rate of Mathieu procedure was 30% (5 / 15), the complication rate of TPIF procedure was 42.9% (233 / 543), the complication rate of TPIF+Duplay procedure was 44.4% (20 / 45), the complication rate of Koyanagi procedure was 45.6% (26 / 57), and the complication rate of staged surgery was 17.9% (25 / 140).

[0071] The number of reoperations for children with complications after different surgical procedures is detailed in Table 3. There were 35 cases (7.3%, 35 / 481) with 3 or more unplanned reoperations, of which 14.3% were due to urethral fistula and 85.7% were due to urethral stenosis. Among the children who underwent 6 unplanned reoperations, one had a 3cm urethral defect and a UDR of 0.86 during the operation, who underwent TPIF. After the operation, urethral stenosis occurred. After 2 repeated urethral dilations, urethral fistula was performed, but it was ineffective. After the fistula was closed, urethral stenosis occurred again, so urethral fistula and fistula closure were performed again. Among the children who underwent 5 reoperations, one had a 3cm urethral defect and a UDR of 0.86 during the operation, who underwent TPIF. Urethral fistula repair was performed repeatedly after surgery due to urethral fistula. Details are shown in Table 3. In one case, the urethral defect was 3 cm in length and the UDR value was 1 during surgery. TPIF was performed. After two urethral dilatations, urethrotomy was performed, which was ineffective. After urethrotomy, urethral fistula was closed. Urethral fistula repair was performed again. In another case, the urethral defect was 3.5 cm in length and the UDR value was 0.7 during surgery. TPIF was performed. After three urethral dilatations, urethrotomy and fistula closure were performed.

[0072] Table 3 Number of reoperations in children with complications after different surgical procedures

[0073] Note: MAGPI, Meatal Advancement and Glanuloplasty Incorporated procedure; TIP, tubularized incised plate urethroplasty; Onlay, transverse island flap urethroplasty; Mathieu, meatal base flap urethroplasty; TPIF, transverse preputial island flap urethroplasty; TPIF+Duplay, TPIF combined with in situ flap Thiersch-Duplay procedure; Koyanagi, pedicled preputial flap urethroplasty based on the meatus.

[0074] Construction and application of a clinical prediction model for urethral plate transecting during hypospadias surgery - Construction of a clinical prediction model for urethral plate transecting during hypospadias surgery.

[0075] A comparison of baseline characteristics between the hospital data cohort and the multicenter cohort revealed statistically significant differences between the two groups in age at surgery, penis length, glans penis length, glans penis width, urethral opening location before correction of chordee, and degree of chordee (P < 0.05). Therefore, a clinical prediction model for intraoperative urethral plate transection was constructed using the hospital data cohort as the training set and internally validated. The model was externally validated using the multicenter cohort as the validation set.

[0076] In the hospital data cohort, 369 patients were divided into a urethral plate retention group and a urethral plate transection group, 450 patients, based on whether the urethral plate was retained during surgery. The baseline clinical data of the two groups are shown in Table 4. Statistical analysis revealed that, with the exception of foreskin morphology and superficial foreskin vascularity, the two groups showed statistically significant differences in all other variables (P < 0.05). A univariate logistic regression analysis was performed, using whether the urethral plate was transected intraoperatively as the dependent variable. Preoperative measurable parameters in the hospital data cohort, including penis length (cm), glans penis width (mm), glans penis length (mm), urethral plate width (mm), preoperative penile curvature (°), foreskin morphology, and the position of the urethral opening before correction of penile curvature, were included. Penile length, glans penis length, glans penis width, urethral plate width, and penile curvature were risk factors for intraoperative urethral plate transection (see Table 5 for details). Statistically significant parameters (P < 0.05) were included in a multivariate logistic regression analysis, which revealed that glans penis length, glans penis width, and penile curvature were influencing factors for urethral plate transection (P < 0.05). A collinearity test was performed on these three variables, and their VIFs were all within a range of 1-1.5, indicating the absence of multicollinearity.

[0077] Therefore, glans penis length, glans penis width, and penile curvature were incorporated into the clinical prediction model for urethral plate transecting during hypospadias surgery, and a nomogram was constructed. Figure 1 The Hosmer-Lemeshow goodness-of-fit test was performed on the model, χ 2 =7.317, P=0.503, indicating that the model fit is good. The calibration curve was used to evaluate the consistency of the model, showing that the predicted value of the model for intraoperative transverse urethral plate obtained by the nomogram was consistent with the actual observed value. Figure 2 Bootstrap was used to resample and observe the stability of the regression model. The number of resamples was 1000. The internal validation of the model was performed. The ROC curve was drawn using the "pROC" package. The area under the ROC curve (AUC) (i.e., C index) was calculated to be 0.914 (95% CI: 0.895, 0.932). Figure 3 , and external validation was performed using a multicenter cohort, with an AUC of 0.87 (95% CI: 0.843, 0.897), see Figure 4 , suggesting that the prediction model has a good ability to predict the transverse urethral plate during hypospadias surgery. The DCA curve of the prediction model training set is detailed in Figure 5 ,According to the decision curve, when the threshold of the model is set in the range of 1%-99%, the ,decision curve is above the None line and the All line, within which ,the model has clinical practicality.

[0078] Table 4 Baseline data of children in the two groups of urethral plate retention and transection in the hospital data cohort

[0079]

[0080]

[0081] Note: Continuous variables are expressed as median (25% quantile, 75% quantile) [M(P25, P75)]; categorical variables are expressed as frequency (percentage); UDR, urethral defect ratio;

[0082] Table 5 Analysis of risk factors for urethral plate transection in the hospital data cohort

[0083]

[0084] Application of intraoperative urethral plate transection prediction model:

[0085] In daily clinical work, surgeons can refer to Figure 1 Preoperative scoring of children with hypospadias is performed, i.e., glans penis length (mm), glans penis width (mm), and penis curvature (°) are measured preoperatively, and then the total score is added up according to the score of each variable, and the predicted probability of urethral plate transecting during the child's operation can be obtained by projecting the total score downward. The higher the total score, the greater the probability of urethral plate transecting during the child's operation. In order to facilitate the accurate calculation of the probability of urethral plate transecting during the operation, the present embodiment converts the nomogram into a risk prediction formula, i.e., probability of urethral plate transecting (probability, p) = exp(10.401+0.134* glans penis length-1.091* glans penis width+0.068* penis curvature) / (1+exp(10.401+0.134* glans penis length-1.091* glans penis width+0.068* penis curvature)).

[0086] The clinical data of 1500 children were incorporated into this formula and scored individually. The probability of urethral plate transection was calculated. This probability was used as the independent variable, and whether the urethral plate was transected during surgery was used as the dependent variable. A receiver operating characteristic (ROC) curve was plotted, and the Youden index was calculated. The optimal cutoff value was determined to be 56.7% (sensitivity 78.6%, specificity 84.9%, accuracy 81.6%), with an AUC of 0.895 (95% CI: 0.877, 0.913). Therefore, using 56.7% as the cutoff value, the 1500 children were divided into a severe group and a non-severe group. There were 723 children in the severe group and 777 in the non-severe group. Detailed baseline clinical characteristics of the two groups are shown in Table 6. As shown in Table 6, the glans penis length (10 (9, 12) mm) in the severe hypospadias group was smaller than the non-severe group (12 (10, 14) mm), the glans penis width (13 (12, 13.5) mm) in the severe group was smaller than the non-severe group (14.5 (14, 15.5) mm), and the degree of penile curvature (70 (55, 90)°) in the severe group was significantly greater than 30 (20, 42)°. These three indicators can be used as objective indicators of the severity of hypospadias and the degree of penile development. Statistical comparison between the two groups revealed statistically significant differences in age, penis length, glans penis length and width, superficial preputial vascularity, urethral plate width, urethral opening position before and after correction of penile curvature, urethral defect length, penile curvature, UDR value, and Barcat classification (P < 0.05). Therefore, the predicted probability of 56.7% calculated by the clinical prediction model of intraoperative urethral plate transection can be used as the cutoff value, which can be used as a grouping method to reflect the severity of hypospadias. After this grouping, the surgical procedure selection and postoperative complications of the two groups of children were statistically significant. Therefore, this classification method has good value in predicting the surgical method.

[0087] Table 6 Clinical baseline characteristics of severe and non-severe hypospadias groups

[0088]

[0089]

[0090] Note: Continuous variables are expressed as median (25% quantile, 75% quantile) [M(P25, P75)]; categorical variables are expressed as frequency (percentage); UDR, urethral defect ratio.

[0091] Selection of urethroplasty procedures with urethral plate preservation

[0092] Among 715 patients who underwent urethroplasty with urethral plate preservation, the primary procedures were TIP (69.0%) in 493 cases, Onlay (19.3%) in 138 cases, MAGPI (9.7%), and Mathieu (2.1%). During a follow-up of 42 months (range, 29-61), only 4.3% of patients experienced complications after MAGPI. In summary, in patients who underwent MAGPI, the urethral opening was located near the fossa navicularis in approximately 82.61% of cases after correction of chordee, with a urethral defect length of 0.3 (0.2, 0.4) cm, a chordee angle of 30 (20-50) degrees, and a UDR of 0.08 (0.06, 0.1). Therefore, this study suggests that, with urethral plate preservation, if the chordee is fully corrected and the urethral opening is located distal to the coronal sulcus and near the fossa navicularis, the MAGPI procedure should be performed. Furthermore, the Mathieu procedure currently accounts for only approximately 2.1% of cases at multiple centers in China, and its postoperative complications are approximately 30%, making it a less mainstream procedure. Therefore, in establishing a flowchart for personalized surgical selection strategies for hypospadias, this application will not further analyze the Mathieu procedure, focusing instead on the selection criteria for the TIP and Onlay procedures.

[0093] Univariate logistic regression analysis was performed on the TIP and Onlay groups, with surgical procedure selection as the dependent variable. The results showed that penis length, glans penis length, glans penis width, urethral plate width, urethral defect length, and UDR value were influencing factors for surgical procedure selection (see Table 7). Multivariate logistic regression analysis also showed that penis length and urethral plate width were influencing factors for surgical procedure selection (P < 0.05). The urethral plate width in the TIP group was 5 (4, 6) mm, which was greater than the urethral plate width of 4 (3, 5) mm in the Onlay group. Clinically, urethral plate width has a greater impact on the choice of TIP versus Onlay procedure than penis length. Therefore, in this example, urethral plate width was used as the independent variable and the choice of TIP versus Onlay procedure was used as the dependent variable. Receiver-operating characteristic (ROC) curves were plotted and the Youden index was calculated. The optimal critical value was determined to be 4.25 mm (sensitivity 67%, specificity 66%, accuracy 66%), with an AUC of 0.67 (95% CI: 0.63, 0.71).

[0094] Therefore, 631 cases of children were divided into narrow urethral plate group (260 cases) and wide urethral plate group (371 cases) with 4.25 mm as the dividing value. The proportion of Onaly operation in narrow urethral plate group was 35.38%, which was higher than 12.4% in wide urethral plate group, and the difference was statistically significant (P<0.05). The complication rate in narrow urethral plate group was 31.54%, which was higher than 23.45% in wide urethral plate group, and the difference was statistically significant (P<0.05). Among them, 168 cases in narrow urethral plate group selected TIP operation, 27.98% of the children had complications, 92 cases selected Onlay operation, 38.04% of the children had complications, and the difference in complication rate between the two groups of children was not statistically significant (P=0.095). In the wide urethral plate group, 325 cases selected TIP operation, 21.54% of the children had complications, and 46 cases selected Onlay operation, 36.96% of the children had complications, and the difference in complication rate between the two groups of children was statistically significant (P=0.021). Therefore, it is recommended that in narrow urethral plate group, that is, in children with urethral plate width less than 4.25 mm, TIP operation should be used with caution, and Onlay operation can be selected.

[0095] Table 7 Influence factors of TIP operation and Onlay operation

[0096]

[0097]

[0098] Note: UDR, Urethral Defect Ration.

[0099] Transverse urethral plate urethroplasty operation selection:

[0100] In children with transverse urethral plate urethroplasty, the complication rate of TPIF operation group in this embodiment was 42.9%, the complication rate of TPIF+Duplay was 44.4%, the complication rate of Koyanagi operation was 45.6%, and there was no statistical significance in the complication rates of the three groups (χ 2 =0.182, P=0.913), and the complication rate of the staged operation group was 17.9%. There was a statistically significant difference in the complication rates between the staged operation and the one-stage operation (χ 2 =31.272, P<0.001), but the children in the staged operation group needed at least two operations, and the one-stage operation had the characteristics of high complication rate. The difficulty for the operator is to choose the staged operation or the one-stage operation under the condition of penis anatomy to reduce the number of operations and improve the treatment effect. Therefore, when establishing the individualized operation selection strategy flowchart of hypospadias, this embodiment focuses on the selection of one-stage operation and staged operation after transverse urethral plate.

[0101] Among the 785 patients who underwent urethral plate transecting, 645 underwent a single-stage procedure, and 145 underwent a staged procedure. Baseline clinical data for the two groups are shown in Table 8. The penile anatomy of the patients who underwent staged surgery was inferior to that of the single-stage group, as demonstrated by smaller glans penis width, more severe Barcat classification, longer urethral defect length, larger UDR values, and greater penile curvature. The differences between the two groups were statistically significant (P < 0.05).

[0102] 785 children were divided into two groups according to the number of surgeries required for cure: a single-surgery group and a multiple-surgery group. The multiple-surgery group included children who underwent staged surgery and those who required reoperation due to complications from the first-stage surgery. A univariate logistic regression analysis was performed on the two groups of children, using the urethral defect length, Barcat classification, and UDR value measured after correction of penile curvature as independent variables and the number of surgeries as the dependent variable. The results showed that the urethral defect length and UDR value were influencing factors of the number of surgeries, as shown in Table 9. Statistically significant indicators (P < 0.05) were screened and included in the multivariate logistic regression analysis. The results showed that the urethral defect length was an influencing factor of the number of surgeries (P < 0.05). Therefore, in this example, the urethral defect length was used as the independent variable and the number of surgeries as the dependent variable to draw the ROC curve and calculate the Youden index. The optimal critical value was determined to be 3.55 cm (sensitivity 57%, specificity 73%, accuracy 42%), with an AUC of 0.673 (95% CI: 0.636, 0.71).

[0103] Therefore, the urethral defect length of 3.55 cm was used as the cutoff value, and 785 children were divided into a non-severe group (449 cases) and a severe group (336 cases) after urethral plate transection. The clinical baseline characteristics of the two groups of children are shown in Table 10. It can be seen that the proportion of the severe group using staged surgery (30.36%) was greater than that of the non-severe group (8.46%), and the complication rate of the severe group (45.83%) was greater than that of the non-severe group (33.41%). The difference between the two groups was statistically significant (P<0.05). The total number of surgeries for children in the severe group using one-stage or staged surgery is detailed in the half-violin plot. Figure 2-5 In the severe group, 6.84% of the patients who underwent one-stage surgery had 2 or more reoperations after complications occurred, while 2.94% of the patients who underwent staged surgery had 2 or more reoperations after complications occurred. There was no statistically significant difference between the two groups (χ 2 =2.022, P=0.155). Therefore, it is recommended that in children who require urethral transection for correction of penile chordee and whose urethral defect length is greater than 3.55 cm, a single-stage urethroplasty should be chosen with caution. Staged surgery is recommended to reduce the difficulty of the single-stage surgery, reduce the need for multiple reoperations to deal with complications, and thus reduce the overall complication rate.

[0104] Table 8 Baseline clinical data of children who underwent one-stage and staged surgery after urethral plate transection

[0105]

[0106]

[0107] Note: Continuous variables are expressed as median (25% quantile, 75% quantile) [M(P25, P75)]; categorical variables are expressed as frequency (percentage); UDR, urethral defect ratio.

[0108] Table 9 Analysis of factors affecting the number of surgeries after urethral plate transection

[0109]

[0110] Note: UDR, urethral defect ratio.

[0111] Table 10 Clinical baseline data of the severe group and non-severe group after urethral plate transection

[0112]

[0113]

[0114] Note: Continuous variables are expressed as median (25% quantile, 75% quantile) [M(P25, P75)]; categorical variables are expressed as frequency (percentage); UDR, urethral defect ratio.

[0115] Construction of a flowchart for individualized surgical strategy for hypospadias:

[0116] Based on this embodiment, the first step of the hypospadias surgery selection strategy process is that in clinical work, the surgeon first measures the glans penis length (mm), glans penis width (mm), and penis curvature (°) before performing urethroplasty, and then refers to Figure 1 Children with hypospadias were scored preoperatively. The scores for each variable were summed to obtain a total score. This total score was then projected downward to determine the predicted probability of urethral plate transection during surgery. A higher total score indicates a greater probability of urethral plate transection (the probability of urethral plate transection can also be calculated using a formula). The cutoff value of 56.7% predicted by the clinical prediction model for intraoperative urethral plate transection was used as the reference probability of urethral plate transection. If the probability exceeded this value, urethroplasty with urethral plate transection was recommended; if the probability was lower than this value, urethroplasty with urethral plate preservation was recommended.

[0117] The second step of the process is to determine the position of the urethral opening after correction of penile chordalgia in children with retained urethral plate. If it is located between the coronal sulcus and the navicular fossa, the MAGPI procedure is performed; if it is located proximal to the coronal sulcus, the width of the urethral plate is measured, with the urethral plate width of 4.25mm as the boundary. If it is greater than this boundary, the TIP procedure is selected; if it is less than this boundary, the TIP procedure should be used with caution and the Onlay procedure can be selected. In children with transverse urethral plate, the length of the urethral defect is measured. If the length of the urethral defect is greater than 3.55cm, it is recommended to consider staged surgery; if the length of the urethral defect is less than 3.55cm, one-stage surgery (TPIF procedure, TPIF+Duplay procedure, Koyanagi procedure, etc.) can be considered. Figure 6 As shown, in addition, the TPIF postoperative complication nomogram established in the first part of this embodiment can be combined with the urethral defect length and penile curvature to enter the nomogram to calculate the possibility of complications after TPIF surgery. If the possibility of postoperative complications is significantly greater than the average level of complications of TPIF surgery, staged surgery is performed. If the difference in complications between the two is not significant, one-stage or staged surgery can be performed. Therefore, the flowchart of the individualized surgical selection strategy for hypospadias based on the penile anatomical characteristics of multiple centers in my country, which was initially established in this embodiment, is detailed in Figure 7 Based on this embodiment, the first step of the hypospadias surgery selection strategy process is that in clinical work, the surgeon first measures the glans penis length (mm), glans penis width (mm), and penis curvature (°) before performing urethroplasty, and then refers to Figure 1 Children with hypospadias were scored preoperatively. The scores for each variable were summed to obtain a total score. This total score was then projected downward to determine the predicted probability of urethral plate transection during surgery. A higher total score indicates a greater probability of urethral plate transection (the probability of urethral plate transection can also be calculated using a formula). The cutoff value of 56.7% predicted by the clinical prediction model for intraoperative urethral plate transection was used as the reference probability of urethral plate transection. If the probability exceeded this value, urethroplasty with urethral plate transection was recommended; if the probability was lower than this value, urethroplasty with urethral plate preservation was recommended.

[0118] The second step of the procedure is to determine the location of the urethral meatus after straightening of the chordee in patients with a preserved urethral plate. If the urethral meatus is located between the coronal sulcus and the scaphoid fossa, the MAGPI procedure is performed. If the urethral meatus is located proximal to the coronal sulcus, the width of the urethral plate is measured. If the width of the urethral plate is greater than 4.25 mm, the TIP procedure is selected. If the width of the urethral plate is less than 4.25 mm, the TIP procedure is not recommended, and the Onlay procedure can be selected. In patients with a transected urethral plate, the length of the urethral defect is measured. If the length of the urethral defect is greater than 3.55 cm, a two-stage procedure is recommended. If the length of the urethral defect is less than 3.55 cm, a one-stage procedure (TPIF, TPIF+Duplay, Koyanagi, etc.) can be considered. In addition, the length of the urethral defect and the degree of chordee can be input into the nomogram of postoperative complications of the TPIF procedure established in the first part of this embodiment to calculate the probability of postoperative complications of the TPIF procedure. If the probability of postoperative complications is significantly greater than the average level of complications of the TPIF procedure, a two-stage procedure is performed. If the difference in complications between the two procedures is not significant, a one-stage or two-stage procedure can be performed.

[0119] Therefore

[0120] Each of the embodiments in the specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between embodiments can be referred to each other. For the system disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method part.

[0121] The principles and implementation modes of the present application are described by using specific examples in this paper. The above description of the embodiments is only to help understand the method of the present application and its core idea; at the same time, for those skilled in the art, according to the idea of the present application, the specific implementation mode and application range will be changed. In summary, the content of the specification should not be understood as a limitation of the present application.

Claims

1. A prediction method based on an individualized hypospadias surgical procedure selection prediction model, characterized in that: include: Step 1: Collect penile anatomical data; Step 2: Combined with the preoperative penile anatomy data, a clinical prediction model for predicting intraoperative urethral plate transection was established; Step 3: Based on the intraoperative penile anatomical data, determine the most appropriate penile anatomy for various surgical procedures in patients with preserved urethral plates, and determine the most appropriate penile anatomy for single-stage and staged procedures in patients with transected urethral plates. Step 4: Integrate steps 2 and 3 to establish a data-based flowchart for individualized surgical selection strategies for children with hypospadias; Step 5: Use statistical software to organize and analyze data, draw a nomogram of the clinical prediction model, perform internal validation, explain the discrimination of the model, test the calibration of the model and draw a calibration curve, evaluate the consistency between the model's predicted risk and the actual risk, and use the clinical decision curve to evaluate the clinical benefit of the model. The height reflects the size of the net benefit, and the width reflects the breadth of the model's scope of application.

2. The prediction method based on the prediction model for individualized hypospadias surgical procedure selection according to claim 1, characterized in that: Inclusion criteria for the described penile anatomical data included: Children with hypospadias who are undergoing first diagnosis and treatment, have complete penile anatomical data, and have complete follow-up information; Exclusion criteria were: hypospadias with intact foreskin, previous history of circumcision or urethra-related surgery, missing key penile anatomical data, chromosomal or genetic abnormalities, combined sexual developmental abnormalities, or incomplete follow-up data.

3. The prediction method based on the prediction model for individualized hypospadias surgical procedure selection according to claim 1, characterized in that: The clinical prediction model was constructed using a hospital data cohort as a training set, and a multicenter cohort was used as a validation set for external validation of the clinical prediction model.

4. The prediction method based on the prediction model for individualized hypospadias surgical procedure selection according to claim 1, characterized in that: The surgical procedures include: MAGPI, TIP, and Onlay procedures.

5. The prediction method based on the prediction model for individualized hypospadias surgical procedure selection according to claim 1, characterized in that: The step 5 is specifically as follows: SPSS 23.0 statistical software was used for data collation and analysis, and the nomogram function of the rms package in R Studio statistical software was used to draw the nomogram of the clinical prediction model: first, a normality test was performed on the continuous variables. The test results showed that age, penis length, glans penis length, glans penis width, urethral plate width, urethral plate width after longitudinal incision, urethral defect length, penile curvature and UDR value did not meet the normal distribution. Then, the variables were described by frequency, and the inter-group comparisons were performed using the chi-square test or Fisher's exact probability method. Continuous variables that did not meet the normal distribution were expressed as medians. The Wilcox test was used to compare the two groups for statistical differences. The presence or absence of complications after hypospadias surgery was used as the dependent variable, and non-parametric tests were used for inter-group comparisons. Spearman correlation was used to analyze continuous variables. When r < 0.4, it was considered a weak correlation, r between 0.4-0.7 was a moderate correlation, and r > 0.7 was a high correlation. The factors with P < 0.05 in the univariate logistic regression analysis were included in the multivariate logistic regression analysis. Multivariate analysis was used to accurately estimate the impact of multiple influencing factors and avoid the influence of confounding factors on the analysis results. The risk factors for complications after hypospadias surgery were analyzed using the logistic regression method, and the odds ratio and 95% confidence interval were calculated. The forward stepwise method was used to select the independent variables that were finally introduced into the prediction model. The variance inflation factor was used to evaluate whether there was multicollinearity in the predictive factors included in the model. The VIF value was greater than 1. The closer it was to 1, the lighter the multicollinearity was, and vice versa. A VIF value less than 10 was within the acceptable range. After the nomogram of the clinical prediction model was established, 1000 Bootstrap self-sampling methods were used for internal validation to further evaluate the effect of the nomogram model. The pROC package was used to draw the ROC curve, and the AUC under ROC was used to explain the discrimination of the model. The AUC value range was 0-1. The closer it was to 1, the better the model's discrimination ability was. The AUC was equal to 0. When AUC is 0.5, it means that the model has no predictive ability; when it is less than 0.5, it means that the model prediction is opposite to the actual result; when AUC is less than 0.6, it is considered that the discrimination is poor; when AUC is 0.6-0.75, it is considered that the model has certain discrimination ability; when AUC is greater than 0.75, it is considered that the discrimination ability is good. The Hosmer-Lemeshow goodness of fit test is used to test the calibration of the prediction model, and a calibration curve is drawn to evaluate the consistency between the model-predicted risk and the actual risk. The closer the calibration line of the model is to the standard line, the better the calibration; the clinical decision curve is used to evaluate the clinical benefit of the model. The height reflects the size of the net benefit, and the width reflects the breadth of the model's application range.

Images