Clinical predictive model to estimate probability of remission in patients with lupus nephritis

Kang, Yingxi; Zuo, Yongdi; He, Manrong; Duo, Lijin; Chen, Xiaolei; Tang, Wanxin

doi:10.1016/j.intimp.2022.108966

Cited by 9 publications

(4 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In adjusted analysis, the potential confounders included age, body mass index (BMI), hemoglobin (Hb), percent of sclerosing glomeruli, histologic activity score, histologic chronicity score, RBC active, ACEI/ARB comedication, dsDNA positive, renal pathology class, and crescent shape pathology were adjusted with multivariable logistic regression. [18][19][20] The baseline urine protein creatinine ratio (UPCR) was adjusted in a multilevel Gaussian regression model of UPCR, assuming treatment impact of the UPCR linearity on the follow-up day.…”

Section: Discussionmentioning

confidence: 99%

The efficacy of induction treatment in Thai patients with lupus nephritis: Observational cohort analysis

Pichaiwong

Lawanaskol²,

Phinyo

et al. 2023

Lupus

View full text Add to dashboard Cite

Background For active proliferative lupus nephritis (LN), intravenous cyclophosphamide (IVCYC) is a standard therapy whereby mycophenolate mofetil (MMF) is less effective than IVCYC, according to the clinical trials in non-Asians. In this observational study, the clinical outcomes will be compared among Thai patients. Methods We had analyzed 93 adult patients who underwent a renal biopsy for active proliferative LN confirmation between January 2013 and June 2021. The assessment of the response outcomes compared the induction treatment of IVCYC versus MMF. The primary endpoint had achieved complete remission (CR) at 24 weeks, while the secondary endpoint is overall remission (OR) and urine protein creatinine ratio (UPCR) changed over time. Results 93 LN patients were separated into two groups; 40 in IVCYC and 53 in MMF. In the 24th week, based on unadjusted analysis, patient had achieved CR 20.0% of IVCYC, whereas 28.3% of MMF had achieved CR. Unadjusted CR Risk difference was −0.08 (95% CI −0.26, −0.09, p-value = 0.351) and the adjusted CR risk difference was −0.19 (95% CI −0.42, 0.04, p-value = 0.098). The unadjusted OR risk difference was −0.06 (95% CI −0.26, 0.14, p-value = 0.553) while adjusted OR risk difference was −0.24 (95% CI −0.50, 0.02, p-value = 0.067). Unadjusted UPCR mean was −0.29 (95% CI −0.77, 0.17, p-value = 0.210) and adjusted UPCR mean was −0.27 (95% CI −0.88, 0.32, p-value = 0.366). Conclusions The induction treatment with either IVCYC or MMF had similar efficacy in Thai LN patients. The decision of treatment should be taken by applying an individualized therapeutic strategy and balancing risks, costs, and benefits.

show abstract

Section: Discussionmentioning

confidence: 99%

The efficacy of induction treatment in Thai patients with lupus nephritis: Observational cohort analysis

Pichaiwong

Lawanaskol²,

Phinyo

et al. 2023

Lupus

View full text Add to dashboard Cite

show abstract

“…Similar model was established in Kang et al's study. 42 In addition to demographic characteristics and laboratory indicators, their article also analyzed the pathology and medication parameters. In order to ensure the integrity of the data, we did not include these indicators.…”

Section: Discussionmentioning

confidence: 99%

Analysis of risk factors and development of a nomogram prediction model for lupus nephritis in systemic lupus erythematosus patients

Zhen

Wang

Qin

et al. 2023

Lupus

View full text Add to dashboard Cite

Background This study aimed to explore risk factors for lupus nephritis (LN) in systemic lupus erythematosus (SLE) patients and establish a Nomogram prediction model based on LASSO-logistic regression. Methods The clinical and laboratory data of SLE patients in Meishan People’s Hospital from July 2012 to December 2021 were analyzed retrospectively. All SLE patients were divided into two groups with or without LN. Risk factors were screened based on LASSO-logistic regression analysis, and a Nomogram prediction model was established. The receiver operating characteristic curve, calibration curves, and decision curve analysis were adopted to evaluate the performance of the Nomogram model. Results A total of 555 SLE patients were enrolled, including 303 SLE patients with LN and 252 SLE patients without LN. LASSO regression and multivariate logistic regression analyses showed that ESR, mucosal ulcer, proteinuria, and hematuria were independent risk factors for LN in SLE patients. The four clinical features were incorporated into the Nomogram prediction model. Results showed that calibration curve was basically close to the diagonal dotted line with slope 1 (ideal prediction case), which proved that the prediction ability of the model was acceptable. In addition, the decision curve analysis showed that the Nomogram prediction model could bring net clinical benefits to patients when the threshold probability was 0.12–0.54. Conclusion Four clinical indicators of ESR, mucosal ulcer, proteinuria, and hematuria were independent risk factors for LN in SLE patients. The predictive power of the Nomogram model based on LASSO-logistic regression was acceptable and could be used to guide clinical work.

show abstract

“…15 As such, deep learning algorithm-based models, using time-series data to predict the risk of LN relapse, could be promising. However, except for some models using traditional methods with data in snapshot, 16 no such models have yet been developed. One of the difficulties in implementing a deep learning algorithm to predict LN relapse risk is that standard deep learning models were designed for predicting binary outcomes.…”

Section: Strengths and Limitations Of This Studymentioning

confidence: 99%

Deep learning model to predict lupus nephritis renal flare based on dynamic multivariable time-series data

Huang,

Chen,

Song

et al. 2024

BMJ Open

View full text Add to dashboard Cite

ObjectivesTo develop an interpretable deep learning model of lupus nephritis (LN) relapse prediction based on dynamic multivariable time-series data.DesignA single-centre, retrospective cohort study in China.SettingA Chinese central tertiary hospital.ParticipantsThe cohort study consisted of 1694 LN patients who had been registered in the Nanjing Glomerulonephritis Registry at the National Clinical Research Center of Kidney Diseases, Jinling Hospital from January 1985 to December 2010.MethodsWe developed a deep learning algorithm to predict LN relapse that consists of 59 features, including demographic, clinical, immunological, pathological and therapeutic characteristics that were collected for baseline analysis. A total of 32 227 data points were collected by the sliding window method and randomly divided into training (80%), validation (10%) and testing sets (10%). We developed a deep learning algorithm-based interpretable multivariable long short-term memory model for LN relapse risk prediction considering censored time-series data based on a cohort of 1694 LN patients. A mixture attention mechanism was deployed to capture variable interactions at different time points for estimating the temporal importance of the variables. Model performance was assessed according to C-index (concordance index).ResultsThe median follow-up time since remission was 4.1 (IQR, 1.7–6.7) years. The interpretable deep learning model based on dynamic multivariable time-series data achieved the best performance, with a C-index of 0.897, among models using only variables at the point of remission or time-variant variables. The importance of urinary protein, serum albumin and serum C3 showed time dependency in the model, that is, their contributions to the risk prediction increased over time.ConclusionsDeep learning algorithms can effectively learn through time-series data to develop a predictive model for LN relapse. The model provides accurate predictions of LN relapse for different renal disease stages, which could be used in clinical practice to guide physicians on the management of LN patients.

show abstract

Clinical predictive model to estimate probability of remission in patients with lupus nephritis

Cited by 9 publications

References 26 publications

The efficacy of induction treatment in Thai patients with lupus nephritis: Observational cohort analysis

The efficacy of induction treatment in Thai patients with lupus nephritis: Observational cohort analysis

Analysis of risk factors and development of a nomogram prediction model for lupus nephritis in systemic lupus erythematosus patients

Deep learning model to predict lupus nephritis renal flare based on dynamic multivariable time-series data

Contact Info

Product

Resources

About