Risk prediction method for intermuscular venous thrombosis of knee arthroplasty patient

Abstract

The invention provides a risk prediction method for intermuscular venous thrombosis of a knee arthroplasty patient. The method comprises the following steps: acquiring potential risk factors of intermuscular venous thrombosis of a research object; carrying out single-factor analysis on the potential risk factors, and screening single-factor variables which are strongly related to the formation of the intermuscular venous thrombosis of the research object; analyzing Plt in the single factor; 0.2 single factor variables are incorporated into multivariable logistic regression analysis, independent risk variables, significantly related to intermuscular venous thrombosis, of a research object are screened, and a risk prediction column graph model of intermuscular venous thrombosis of the research object is constructed; and performing distinction degree, calibration degree and clinical adaptability test on the column graph model. By constructing the inter-muscular venous thrombosis column diagram risk prediction model of the knee joint replacement patient, the occurrence risk of inter-muscular venous thrombosis of the knee joint replacement patient can be specifically evaluated and predicted.

Description

Technical Field

[0001] The present invention relates to the field of biomedical technology, and in particular to a method for predicting the risk of intermuscular vein thrombosis in patients undergoing knee replacement. Background Art

[0002] Knee replacement is currently an effective means of treating knee osteoarthritis in the end stage of the elderly. Among them, deep venous thrombosis (DVT) of the lower limbs is one of the most serious complications after knee replacement.

[0003] DVT is a peripheral vascular disease that affects the veins of the lower extremities. It increases the risk of multiple complications such as lower extremity swelling, secondary varicose veins, stasis ulcers, pulmonary embolism, and long-term deep vein insufficiency, which is not conducive to the patient's postoperative recovery. Studies have found that the incidence of DVT after knee replacement surgery is 11.5%, of which muscular calf vein thrombosis (MCVT) accounts for 75%.

[0004] MCVT usually refers to thrombosis that originates from the gastrocnemius vein, soleus vein, and related communicating branches. It is a type of deep vein thrombosis in the lower leg or isolated distal deep vein thrombosis in the lower limb. Patients after knee replacement often experience hypercoagulable state, blood stasis, and venous wall damage. At the same time, patients are often elderly and have high-risk factors such as postoperative blood loss, pain, and bed rest, which can easily lead to the formation of MCVT. If MCVT is not treated in time, it can lead to other serious complications.

[0005] Most calf MCVT is primary because it does not affect blood return and has a small range at the onset. The inflammatory response it stimulates is mild. Most patients lack specific manifestations clinically. Pain is a common symptom, but 50%-60% of patients may still have no clinical symptoms and are easily missed. Therefore, it is unreliable and inaccurate to assess the incidence of CRT in missed patients based solely on clinical symptoms, which may lead to a high missed diagnosis rate of MCVT. It is necessary to assess the possibility of MCVT in patients based on their symptoms, signs, and possible risk factors.

[0006] Currently, the Wells-DVT scoring scale is the most commonly used clinical probability scoring scale for DVT. However, studies have shown that the risk factors for MCVT, which is a peripheral DVT, are different from those for the formation of proximal DVT in the lower limbs. It is more likely to be related to transient risk factors (such as recent hospitalization, recent surgery, recent travel, and varicose veins in the lower legs, etc.), while chronic high-risk factors (such as advanced age, combined cardiovascular disease, blood system diseases, rheumatoid arthritis, secondary joint replacement, etc.) are prone to central DVT formation. Therefore, relying solely on the Wells-DVT scoring scale may underestimate the possibility and cannot accurately assess the possibility of MCVT formation in patients. Summary of the invention

[0007] In order to solve the above technical problems, the present invention provides a method for predicting the risk of intermuscular vein thrombosis in patients undergoing knee replacement, the method comprising the following steps: S1. Obtain the potential risk factors for intermuscular vein thrombosis in the subjects; S2. Performing univariate analysis on the potential risk factors to screen for univariate variables that are strongly correlated with the intermuscular vein thrombosis in the research subjects; S3. Incorporating the univariate variables with P<0.2 in the univariate analysis into a multivariate logistic regression analysis, screening the independent risk variables significantly associated with the intermuscular vein thrombosis in the research subjects based on the results of the multivariate logistic regression analysis, and constructing a risk prediction nomogram model for intermuscular vein thrombosis in the research subjects; S4. Test the discrimination, calibration and clinical adaptability of the nomogram model.

[0008] Furthermore, the potential risk factors include demographic characteristics, physiological and genetic characteristics, comorbidities, knee replacement information, hematological test indicators and other clinical characteristics.

[0009] Furthermore, the demographic characteristics include gender and age; The physiological and genetic characteristics include blood type and genetic testing related indicators; wherein the blood type includes O type and non-O type; the genetic testing related indicators include MTHFR and PAI; The comorbidities include hypertension, diabetes, heart disease, and cerebrovascular disease; The knee replacement information includes the side of the knee replacement, bed rest > 72 hours, surgical site type, laryngeal mask general anesthesia, endotracheal anesthesia, peripheral nerve block, intraoperative blood loss, and tourniquet pressure value; wherein the surgical site type includes unicompartmental and total knee; The hematological test indicators include the preoperative and postoperative initial PLT, MPV, ALB, PT, INR, Fg, APTT, D-dimer and CRP values; Other clinical features described included preoperative and postoperative lower extremity swelling.

[0010] Furthermore, the S1 further includes: Calculate the incidence of intermuscular vein thrombosis in the research subjects; The research subjects were randomly divided into two groups of data: a training set and a validation set, wherein the training set was used to construct the nomogram model, and the validation set was used to evaluate the performance of the nomogram model.

[0011] Furthermore, the univariate variables with P < 0.2 in the univariate analysis included: age, blood type, cerebrovascular disease, bed rest > 72 hours, postoperative lower limb swelling, peripheral nerve block, preoperative initial D-dimer value, preoperative initial PT value, preoperative initial CRP value and postoperative initial D-dimer value.

[0012] Furthermore, the independent risk variables include: age, cerebrovascular disease, bed rest > 72 hours, postoperative lower limb swelling, peripheral nerve block, preoperative initial PT value and postoperative initial D-dimer value, wherein the age is over 60 years old.

[0013] Furthermore, the S3 specifically includes: The intermuscular vein thrombosis was used as the target variable and the independent risk variables were used as predictors, and the glm() function was used to fit a logistic regression model.

[0014] Furthermore, the S4 specifically includes: The discrimination ability of the nomogram model is evaluated by drawing a ROC curve and calculating an AUC value; wherein the closer the AUC value is to 1, the higher the discrimination of the nomogram model is; The calibration of the nomogram model was tested by the Hosmer-Lemeshow goodness of fit test; The net advantage of the nomogram model at different threshold probabilities was evaluated by DCA.

[0015] Furthermore, the AUC value of the training set is 0.719, wherein the 95% confidence interval is 0.651-0.788; the AUC value of the validation set is 0.783, wherein the 95% confidence interval is: 0.694-0.873; The P value of the training set is 0.8982, X 2 The value is 3.5126; the P value of the validation set is 0.3922, X 2The value is 8.4342, indicating that the logistic regression model fits well; The training set showed a net advantage in the threshold probability range of 6%-70%; the validation set showed a net advantage in the threshold probability interval of 8%-58%.

[0016] Compared with the prior art, the present invention has the following beneficial effects: The present invention can specifically evaluate and predict the risk of intermuscular vein thrombosis in patients undergoing knee replacement by constructing a nomogram risk prediction model for intermuscular vein thrombosis in patients undergoing knee replacement. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a flowchart of the overall process disclosed in the embodiment of the present invention; Figure 2 A schematic diagram of a risk prediction nomogram model for intermuscular vein thrombosis in patients undergoing knee replacement surgery disclosed in an embodiment of the present invention; Figure 3 A statistical diagram of the discrimination of the test nomogram model disclosed in the embodiment of the present invention; Figure 4 A calibration statistical diagram of the test nomogram model disclosed in an embodiment of the present invention; Figure 5 This is a statistical graph of the clinical adaptability of the test nomogram model disclosed in the embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to make the technical scheme and technical effect of the present invention clearer, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments.

[0019] The present invention aims to provide a method for predicting the risk of intermuscular vein thrombosis in patients undergoing knee replacement. By establishing a risk prediction nomogram model for intermuscular vein thrombosis in patients undergoing knee replacement, the risk of intermuscular vein thrombosis in patients undergoing knee replacement can be specifically evaluated and predicted.

[0020] See also Figure 1 The risk prediction method for intermuscular vein thrombosis in patients undergoing knee replacement mainly includes the following steps: Step 1: Data Collection This example retrospectively collected demographic and clinical data of patients who underwent knee replacement surgery in the Department of Orthopedics, First Affiliated Hospital, University of Science and Technology of China from September 1, 2022 to December 31, 2022.

[0021] The inclusion criteria for the study subjects are as follows: (1) Age >18 years old; (2) receiving knee replacement surgery; (3) Preoperative venous ultrasound imaging confirmed the absence of MCVT; (2) Postoperative venous ultrasound imaging confirmed the formation of MCVT.

[0022] Exclusion criteria were as follows: (1) No venous ultrasound imaging examination was performed before and after surgery to screen for CMVT formation; (2) The missing values ​​of potential risk factor variables in the clinical data collection exceed 20%.

[0023] Potential risk factors for intermuscular vein thrombosis in patients undergoing knee arthroplasty were identified through literature review and clinical expertise, and relevant data were collected from the hospital's electronic health information system (HIS).

[0024] In a further embodiment of the present invention, the potential risk factors for intermuscular vein thrombosis of the research subjects include demographic characteristics, physiological and genetic characteristics, comorbidities, knee replacement information, hematological test indicators and other clinical characteristics.

[0025] Specific: Demographic characteristics include gender and age.

[0026] Physiological and genetic characteristics include blood type and genetic testing related indicators; among them, blood type includes O type and non-O type; genetic testing related indicators include MTHFR and PAI, etc.

[0027] Comorbidities include hypertension, diabetes, heart disease, and cerebrovascular disease.

[0028] Knee replacement information included side of knee replacement, bed rest >72 hours, surgical site type, laryngeal mask anesthesia, endotracheal anesthesia, peripheral nerve block, intraoperative blood loss, and tourniquet pressure value; surgical site type included unicompartmental and total knee.

[0029] Hematological examination indicators included preoperative and postoperative initial PLT, MPV, ALB, PT, INR, Fg, APTT, D-dimer, and CRP values.

[0030] Other clinical features include preoperative and postoperative lower extremity swelling.

[0031] The incidence of intermuscular vein thrombosis in the subjects was calculated.

[0032] The research subjects of this example included 103 patients with intermuscular vein thrombosis and 298 patients without intermuscular vein thrombosis, and the incidence of intermuscular vein thrombosis in the sample was 25.69%.

[0033] The research subjects were randomly divided into two groups of data: a training set and a validation set. The training set was used to construct the nomogram model, and the validation set was used to evaluate the performance of the nomogram model.

[0034] Optionally, the subjects are randomly divided into two groups in a ratio of 7:3; the number of samples in the training set is 280; the number of samples in the validation set is 121.

[0035] Step 2: Univariate analysis Univariate analysis was performed on potential risk factors to screen univariate variables that were strongly correlated with intramuscular vein thrombosis in the study subjects.

[0036] Univariate analysis can be performed using one or more of the following: t-test, Mann-Whitney U test, Pearson chi-square test, continuity-corrected chi-square test, or Fisher's exact test.

[0037] Among them, the univariate variables with P < 0.2 in the univariate analysis included: age, blood type, cerebrovascular disease, bed rest > 72 hours, postoperative lower limb swelling, peripheral nerve block, preoperative initial D-dimer value, preoperative initial PT value, preoperative initial CRP value, and postoperative initial D-dimer value.

[0038] It should be noted that the univariate analysis in this study uses a P value less than 0.2 as a looser standard for inclusion in the logistic regression multivariate analysis, based on the differences between multivariate and univariate analysis, avoiding the omission of important information, the balance between statistical significance and practical significance, and flexibility and robustness. This is to ensure that the model can more comprehensively capture the relationship between the independent variable and the dependent variable, thereby improving the accuracy and predictive ability of the model.

[0039] Step 3: Construct a nomogram model The univariate variables with P < 0.2 in the univariate analysis were included in the multivariate logistic regression analysis. Based on the results of the multivariate logistic regression analysis, the independent risk variables significantly associated with the intermuscular vein thrombosis in the research subjects were screened, and a risk prediction nomogram model for intermuscular vein thrombosis in the research subjects was constructed.

[0040] In a further embodiment of the present invention, intermuscular vein thrombosis is used as the target variable and the independent risk variable is used as the predictor, and the glm() function is used to fit the logistic regression model. After the logistic regression model is fitted, the nomogram is drawn using the R software rms package and regplot package.

[0041] Among them, independent risk variables include: age, cerebrovascular disease, bed rest > 72 hours, postoperative lower limb swelling, peripheral nerve block, preoperative initial PT value and postoperative initial D-dimer value, among which the age is over 60 years old and above.

[0042] Specific: Age [60-74 (y)] (OR=1.994, 95%CI:0.898-4.850); Age [≥75 (y)] (OR=2.607, 95%CI:0.955-7.428).

[0043] Cerebrovascular disease (OR=2.593, 95%CI:0.932-6.953).

[0044] Bed rest >72 hours (OR=2.22, 95%CI:1.112-4.387).

[0045] Postoperative lower limb swelling (OR=1.955, 95%CI:0.915-4.083).

[0046] Peripheral nerve block (OR=1.766, 95%CI=0.953-3.266).

[0047] Preoperative initial blood test PT value (OR=0.528, 95%CI:3.10-0.870).

[0048] Initial D-dimer after surgery (OR=1.122, 95%CI:1.043-1.210).

[0049] See also Figure 2 , based on the above 7 independent risk variables, a nomogram model was further constructed to predict the risk of intermuscular vein thrombosis in patients undergoing knee replacement.

[0050] Step 4: Verify the nomogram model The discrimination, calibration and clinical adaptability of the nomogram model were tested.

[0051] See also Figure 3 The discrimination ability of the nomogram model was evaluated by drawing the ROC curve and calculating the AUC value.

[0052] Among them, the closer the AUC value is to 1, the higher the discrimination of the nomogram model is.

[0053] In this scheme, the AUC value of the training set is 0.719, of which the 95% confidence interval is 0.651-0.788; the AUC value of the validation set is 0.783, of which the 95% confidence interval is: 0.694-0.873. According to the fact that the performance of the nomogram model on the validation set is similar to or better than that on the training set, it can be shown that the nomogram model has a certain generalization ability, that is, it can also perform well on new data.

[0054] See also Figure 4 The calibration of the nomogram model was tested by the Hosmer-Lemeshow goodness-of-fit test.

[0055] In this scheme, the P value of the training set is 0.8982, X 2 The value is 3.5126; the P value of the validation set is 0.3922, X 2 The value is 8.4342. According to the P values ​​of the two data sets, both are greater than 0.05, which indicates that the probability predicted by the nomogram model is consistent with the actual observed results, that is, the model has good calibration in both the training set and the validation set.

[0056] See also Figure 5 , the net advantage of the nomogram model at different threshold probabilities was evaluated by DCA.

[0057] In this scheme, the training set showed a net advantage in the threshold probability range of 6%-70%, and the validation set showed a net advantage in the threshold probability range of 8%-58%. That is, the nomogram model showed strong discriminative ability in both sets of data analyzed.

[0058] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for predicting the risk of intermuscular vein thrombosis in patients undergoing knee replacement, characterized in that: The method comprises the following steps: S1. Obtain the potential risk factors for intermuscular vein thrombosis in the subjects; S2. Performing univariate analysis on the potential risk factors to screen for univariate variables that are strongly correlated with the intermuscular vein thrombosis in the research subjects; S3. Incorporating the univariate variables with P<0.2 in the univariate analysis into a multivariate logistic regression analysis, screening the independent risk variables significantly associated with the intermuscular vein thrombosis in the research subjects based on the results of the multivariate logistic regression analysis, and constructing a risk prediction nomogram model for intermuscular vein thrombosis in the research subjects; S4. Test the discrimination, calibration and clinical adaptability of the nomogram model.

2. The method for predicting the risk of intermuscular vein thrombosis in patients undergoing knee replacement according to claim 1, characterized in that: The potential risk factors include demographic characteristics, physical and genetic characteristics, comorbidities, knee replacement information, hematological test indicators, and other clinical characteristics.

3. The method for predicting the risk of intermuscular vein thrombosis in patients undergoing knee replacement according to claim 2, characterized in that: The demographic characteristics include gender, age; The physiological and genetic characteristics include blood type and genetic testing related indicators; wherein the blood type includes O type and non-O type; the genetic testing related indicators include MTHFR and PAI; The comorbidities include hypertension, diabetes, heart disease, and cerebrovascular disease; The knee replacement information includes the side of the knee replacement, bed rest > 72 hours, surgical site type, laryngeal mask general anesthesia, endotracheal anesthesia, peripheral nerve block, intraoperative blood loss, and tourniquet pressure value; wherein the surgical site type includes unicompartmental and total knee; The hematological test indicators include the preoperative and postoperative initial PLT, MPV, ALB, PT, INR, Fg, APTT, D-dimer and CRP values; Other clinical features described included preoperative and postoperative lower extremity swelling.

4. The method for predicting the risk of intermuscular vein thrombosis in patients undergoing knee replacement according to claim 1, characterized in that: The S1 further comprises: Calculate the incidence of intermuscular vein thrombosis in the research subjects; The research subjects were randomly divided into two groups of data: a training set and a validation set, wherein the training set was used to construct the nomogram model, and the validation set was used to evaluate the performance of the nomogram model.

5. The method for predicting the risk of intermuscular vein thrombosis in patients undergoing knee replacement according to claim 3, characterized in that: The univariate variables with P < 0.2 in the univariate analysis included: age, blood type, cerebrovascular disease, bed rest > 72 hours, postoperative lower limb swelling, peripheral nerve block, preoperative initial D-dimer value, preoperative initial PT value, preoperative initial CRP value and postoperative initial D-dimer value.

6. The method for predicting the risk of intermuscular vein thrombosis in patients undergoing knee replacement according to claim 5, characterized in that: The independent risk variables include: age, cerebrovascular disease, bed rest > 72 hours, postoperative lower limb swelling, peripheral nerve block, preoperative initial PT value and postoperative initial D-dimer value, wherein the age is over 60 years old and above.

7. The method for predicting the risk of intermuscular vein thrombosis in patients undergoing knee replacement according to claim 3, characterized in that: The S3 specifically includes: The intermuscular vein thrombosis was used as the target variable and the independent risk variables were used as predictors, and the glm() function was used to fit a logistic regression model.

8. The method for predicting the risk of intermuscular vein thrombosis in patients undergoing knee replacement according to claim 7, characterized in that: The S4 specifically includes: The discrimination ability of the nomogram model is evaluated by drawing a ROC curve and calculating an AUC value; wherein the closer the AUC value is to 1, the higher the discrimination of the nomogram model is; The calibration of the nomogram model was tested by the Hosmer-Lemeshow goodness of fit test; The net advantage of the nomogram model at different threshold probabilities was evaluated by DCA.

9. The method for predicting the risk of intermuscular vein thrombosis in patients undergoing knee replacement according to claim 8, characterized in that: The AUC value of the training set is 0.719, with a 95% confidence interval of 0.651-0.788; the AUC value of the validation set is 0.783, with a 95% confidence interval of 0.694-0.873; The P value of the training set is 0.8982, X 2 The value is 3.5126; the P value of the validation set is 0.3922, X 2 The value is 8.4342, indicating that the logistic regression model fits well; The training set showed a net advantage in the threshold probability range of 6%-70%; the validation set showed a net advantage in the threshold probability interval of 8%-58%.

Images