| Background:As a field of dynamic advancement,arthroscopic techniques are extensively applied in the diagnosis and treatment of a number of knee disorders,which represent one of the common surgical procedures performed by orthopaedic surgeons.Relative to conventional open surgery,the advantages of arthroscopy are as follows:small incision,rapid healing,and rapid postoperative recovery,so it has the characteristics of high safety and less trauma.Following the innovation of arthroscopic technology and the update of minimally invasive instruments and materials,the type of surgery has progressively transitioned from the initial“simple surgery”for examination,cleaning and removal to the treatment of comparatively sophisticated or severe knee diseases.Nevertheless,“minimally invasive”does not necessarily denote“noninvasive”.With the increase in surgical volume and complexity of surgical techniques,the types and incidence of postoperative complications have increased,while the prior studies have primarily concentrated on the more severe but comparatively rare complications after knee arthroscopy,such as deep vein thrombosis and joint cavity infection.In this context,insufficient attention has been paid to the more common acute and chronic pain complications that occur after knee arthroscopy and the results of postoperative pain management have not been satisfactory.If severe and prolonged pain occurs after arthroscopy,it is likely to result in delayed recovery,knee dysfunction,worsening pain-related anxiety or depression,which in turn leads to reduced quality of life and increased financial burden for patients.Nevertheless,clinical studies on post-arthroscopic pain(such as occurrence,outcome,consequences,and influencing factors)have not been systematic to date,in addition to the fact that the research findings are not uniform,making the scientific and standardized management of post-arthroscopic pain still a challenging and evolving field.For this reason,this study will probe the risk factors inducing acute and chronic pain after knee arthroscopy,construct a predictive model and validate the model to assist physicians in selecting and implementing personalized post-arthroscopic analgesic regimens for patients,thereby furnishing a clinical reference for reducing post-operative pain.Objective:1.By analyzing the clinical data collected from patients undergoing knee arthroscopy,the independent risk factors leading to postoperative acute and chronic pain were selected.2.Use the clinical indicators of selected patients to establish a histogram prediction model,and test and evaluate it to evaluate the prediction effect of the model and to provide a basis for screening high-risk groups and implementing personalised protocols.Methods:According to the inclusion and exclusion criteria,patients who were treated in the Department of Sports Medicine of the Second Affiliated Hospital of the Air Force Military Medical University from September 2019 to December 2021 were selected as the subjects of knee arthroscopic surgery.When consulting medical records and guiding patients to fill in questionnaires,the patient’s personal data were:age,gender,surgical side,body mass index(BMI),preoperative pain number score(NRS),injury mechanism,course of disease,smoking history,diabetes history,anxiety,depression,sleep quality,classification of external bridge of articular cartilage injury,and surgical methods,Operation time and flushing volume of joint cavity during operation.Reviewing the previous literature,postoperative acute pain(APSP)was defined as NRS≥4 points within 72 hours after operation,and chronic postoperative pain(CPSP)was defined as pain in the operating area more than 3 months after operation,excluding other factors,and NRS≥4 points within 3months after operation.Follow up and record the pain of patients 72 hours and 3 months after operation,and then divide the patients into two groups,and compare the clinical data of the two groups.Single factor logistic regression analysis was used to select the risk factors with differences,and then multiple factor logistic regression analysis was used to select the indicators of patients with differences,and the independent risk factors of acute and chronic postoperative pain were selected.Apply the selected independent risk factors to the R language to establish a histogram prediction model,use the self-help sampling method for internal validation,and evaluate the prediction efficiency of the model through the subject work characteristic curve(ROC)and calibration curve.Results:(Part1)A total of 260 patients who underwent arthroscopic knee surgery participated in the study according to the inclusion and exclusion criteria,148 males and 112 females,aged between 16～77,with an average age of(41.61±14.23).When patients were divided according to their pain status 72 hours after surgery,102(39.2%)patients in the postoperative APSP group and 158(60.8%)patients in the non-APSP group.A multifactorial analysis for risk factors associated with contributing to postoperative APSP yielded:age[OR=1.063,95%CI(1.028～1.100),P<0.05],BMI≥24kg/m~2[OR=2.104,95%CI(1.078～4.109),P<0.05],preoperative NRS score≥4[OR=3.547,95%CI(1.781～7.063),P<0.05],anxiety score≥9[OR=3.311,95%CI(1.106～9.912),P<0.05],depression score≥9[OR=2.892,95%CI(1.044～8.012),P<0.05],sleep quality score≥5[OR=3.030,95%CI(1.332～6.893),P<0.05],cruciate ligament reconstruction or intercondylar spine avulsion fracture reduction surgical approach[OR=3.883,95%CI(1.666～9.048),P<0.05],and Outerbridge classification grade III-IV[OR=9.038,95%CI(2.031～40.209),P<0.05]was an independent risk factor for APSP after knee arthroscopy.Based on the results of the multifactorial analysis,a column line graph prediction model was established,and the Boot Strap self-sampling method yielded a Concordance index(C-index)of 0.831 and an area under the ROC curve of 0.852,95%CI(0.807～0.896),with good agreement between the predicted results of the calibration curve and the actual observed results.The above results indicate that the model has high predictive efficacy.(Part 2)The patients in the first part of this study continued to receive follow-up,and no one missed the visit.This group was based on the pain of patients 3 months after operation.There were 36 cases(13.8%)in the CPSP group and 224 cases(86.2%)in the non-CPSP group.A multifactorial analysis of risk factors associated with contributing to postoperative CPSP showed that:age[OR=1.091,95%CI(1.044～1.140),P<0.05],BMI≥24kg/m~2[OR=3.263,95%CI(1.130～9.422),P<0.05],Outerbridge classification grade III-IV[OR=14.091,95%CI(2.834～70.053),P<0.05],and NRS≥4 points at 72hours postoperatively[OR=3.846,95%CI(1.433～10.326),P<0.05]were independent risk factors for CPSP after knee arthroscopy.Based on the results of the multifactor analysis to establish a column line graph prediction model,the Boot Strap self-sampling method yielded a C-index of 0.881 and an area under the ROC curve of 0.882,95%CI(0.788～0.960),and the prediction results of the calibration curve were in good agreement with the actual observation.The above results indicate that the model has high predictive efficacy and clinical application value The above results indicate that the model has high predictive efficacy.Conclusion:The incidence rate of APSP after knee arthroscopy was 39.2%,and the incidence rate of CPSP was 13.8%.Different from the previous opinion of orthopedic doctors,this may be an important medical and health problem.Age,BMI≥24kg/m~2,severe postoperative pain with anxiety,depression,sleep disorder,Outerbridge grade III-IV,cruciate ligament reconstruction or surgical reduction of intercondylar spine avulsion fracture are independent risk factors for postoperative APSP.Age,BMI≥24 kg/m~2,Outerbridge grade III-IV and postoperative NRS≥4 at 72 hours were independent risk factors for postoperative CPSP.The prediction model of acute and chronic pain after knee arthroscopic surgery based on the above independent risk factors has high accuracy and specificity,which is conducive to the selection of high-risk patients to implement personalized clinical analgesia decisions,and is currently a valuable model for predicting pain after knee arthroscopic surgery. |
