Risk Factors Analysis And Nomogram Predictive Model Construction Of IgA Nephropathy

BackgroundIgA nephropathy（IgAN）is one of the most common forms of primary glomerular disease today,accounting for 45.3% of primary glomerular disease,particularly in Asia.End stage renal disease（ESRD）,requiring renal replacement therapy.Therefore,early identification of risk factors affecting the prognosis of IgAN and the development of predictive models are important in guiding early intervention and determining the prognosis of IgAN patients.In recent years,a number of indicators have been used to build predictive models for IgAN prognosis.The benefit of immunosuppression in the treatment of IgAN patients is still controversial,and the predictive models of previous studies have rarely included immunosuppression use.ObjectiveTo explore the factors affecting the prognosis and develop a nomogram model to predict the prognosis of IgAN patients.MethodsPatients diagnosed with IgAN by renal biopsy for the first time in our hospital from July 1,2018 to October 1,2021 were used as the study population.Demographic data（gender,age）,clinical laboratory indices（blood pressure,serum albumin,serum uric acid,blood urea nitrogen,blood creatinine,e GFR,24-hour urine protein quantification,urine red blood cell count）,pathological findings（Oxford pathological staging）and whether immunosuppressive drugs were applied(including glucocorticoids alone,cytotoxic drugs alone and glucocorticoids combined with cytotoxic drugs Patients were followed up until the endpoint event,defined as reached remission of urinary protein（both complete and partial remission）.Patients were grouped according to whether they reached the endpoint event and divided into the urinary protein remission group and the non-remission group.Baseline data of patients in the two groups were compared,and indicators with statistically significant differences between the two groups and those considered to be clinically relevant to the prognosis of IgAN patients（ACEI/ARB class drug application,haematuria）were included in the multi-factor Cox regression,and factors affecting the prognosis of IgAN were screened using stepwise regression analysis.The R language was used to construct the columnar line graph model,the model was validated,the consistency index and calibration curve were used to assess the discrimination and accuracy of the model,and the ROC curve was plotted to assess the ability of the nomogram predictive model to predict the prognosis of IgAN patients.Results1.Analysis of influencing factors: A total of 254 patients were included in the study with a follow-up period of 15.00（5.75,20.00）months.At the end of follow-up 144 patients reached remission of urinary protein（both complete and partial remission）and 99 patients had no remission.Comparison of baseline data between the remission and non-remission groups showed statistically significant differences between the two groups in immunosuppression,diastolic blood pressure,e GFR,blood urea nitrogen,blood uric acid,24-hour urine protein quantification,serum albumin and pathological findings（P<0.05）.Multi-factorial Cox regression analysis showed baseline 24-hour urine protein quantification（HR=0.716,95%CI:0.599-0.856,P<0.001）,immunosuppression（HR=0.154,95%CI:0.443-0.931,P=0.020）,renal tubular atrophy/interstitial fibrosis >50%（HR=0.153,95% CI: 0.047-0.497,P=0.002）were not beneficial for urinary protein remission in IgAN patients and were independent factors influencing poor prognosis in IgAN patients.2.Nomogram prediction model: The score of each factor is the score of the nomogram corresponding to the score above,and the results indicate that the higher the baseline 24-hour urine protein quantification,the greater the degree of renal tubular atrophy/interstitial fibrosis,and the use of immunosuppressive agents A lower score on the corresponding nomogram prediction model indicates a lower probability of urinary protein remission.After verification,the nomogram prediction model showed that the consistency index was 0.730（95%CI: 0.683-0.777,P < 0.001）,indicating that the model had a good degree of discrimination.The calibration curve showed that the probability of predicting urinary protein remission in IgAN patients was proportional to the actual probability There is good consistency between the models,indicating that the accuracy of the model is better.The area under the ROC curve was 0.886（95%CI: 0.839-0.923,P<0.001）,indicating that the nomogram prediction model has good predictive ability.ConclusionsBaseline 24-hour urine protein quantification,use of immunosuppressants,and renal tubular atrophy/interstitial fibrosis >50% are adverse factors for the remission of urinary protein in patients with IgAN.The nomogram model established based on the above factors can help predict the prognosis of patients with IgAN.