A case of the syndrome of inappropriate secretion of antidiuretic hormone (SIADH) associated with lupus erythematosus in the central nervous system

Machine learning: Predicting hospital length of stay in patients admitted for lupus flares

BACKGROUND

Although rare, severe systemic lupus erythematosus (SLE) flares requiring hospitalization account for most of the direct costs of SLE care. New machine learning (ML) methods may optimize lupus care by predicting which patients will have a prolonged hospital length of stay (LOS). Our study uses a machine learning approach to predict the LOS in patients admitted for lupus flares and assesses which features prolong LOS.

METHODS

Our study sampled 5831 patients admitted for lupus flares from the National Inpatient Sample Database 2016-2018 and collected 90 demographics and comorbidity features. Four machine learning (ML) models were built (XGBoost, Linear Support Vector Machines, K Nearest Neighbors, and Logistic Regression) to predict LOS, and their performance was evaluated using multiple metrics, including accuracy, receiver operator area under the curve (ROC-AUC), precision-recall area under the curve (PR- AUC), and F1-score. Using the highest-performing model (XGBoost), we assessed the feature importance of our input features using Shapley value explanations (SHAP) to rank their impact on LOS.

RESULTS

Our XGB model performed the best with a ROC-AUC of 0.87, PR-AUC of 0.61, an F1 score of 0.56, and an accuracy of 95%. The features with the most significant impact on the model were "the need for a central line," "acute dialysis," and "acute renal failure." Other top features include those related to renal and infectious comorbidities.

CONCLUSION

Our results were consistent with the established literature and showed promise in ML over traditional methods of predictive analyses, even with rare rheumatic events such as lupus flare hospitalizations.

Value of Autostrain LV in the study of left ventricular systolic function and synchronization in patients with systemic lupus erythematosus

BACKGROUND

There are relatively few studies investigating cardiac structural and functional abnormalities associated with systemic lupus erythematous (SLE). The long-term prognosis of SLE patients is closely related to the cardiovascular events caused by SLE. Accordingly, it is necessary to assess early myocardial systolic function and synchrony.

METHODS

Overall, 90 patients with SLE were randomly selected from our outpatient and inpatient clinics and divided according to SLE Disease Activity Index (SLE-DAI-2000) scores: group A, stable (scores 0-4); group B, mildly active stage (scores 5-9); and group C, moderately active stage (scores ≥10). Each group included 30 patients. Further, 30 sex- and age-matched healthy individuals who were referred for check-ups at the same period were selected as controls (group D). The minimum age for entry into the group was 17 years old. Autostrain LV and three-dimensional quantitative analysis (3DQA) were applied to obtain left ventricular systolic function parameters, information on strain parameters, and correlations between parameters. Simultaneity parameters measured by Autostrain LV and 3DQA were tested for reproducibility.

RESULTS

A two-by-two comparison of groups A-C showed that as the disease activity score increased, AP4LS%, AP2LS%, AP3LS%, and the LV mean overall longitudinal strain all gradually decreased, while LV longitudinal strain peak time standard deviation (Tls-SD) gradually increased, with all differences being statistically significant (p < .05). In groups A-C compared with controls, Tmsv-17-SD, Tmsv-17-Dif, Tmsv-17-SD%, and Tmsv-17-Dif% were all significantly prolonged (p < .05). Further, Tls-SD was positively correlated with Tmsv-17-SD and Tmsv-17-Dif, and there was good agreement between Autostrain and 3DQA for the measurement of left ventricular synchrony indexes, with Tmsv-17-Dif having the best repeatability (intraobserver interclass correlation coefficient (ICC) = .979; interobserver ICC = .848, p < .01).

CONCLUSION

Autostrain LV can accurately detect changes in left ventricular myocardial strain in patients with SLE early in the disease, with simple operation. The 3DQA technique can quantitatively evaluate left ventricular systolic synchronization in patients with SLE, and Autostrain LV synchronization index measurements correlate significantly with 3DQA. Both methods are reproducible, but 3DQA is more sensitive to left ventricular synchronous motion changes.

Systemic lupus erythematosus presenting as hyponatremia-associated rhabdomyolysis: A case report

RATIONALE

Systemic lupus erythematosus (SLE) is an autoimmune disease that involves multiple organs and causes various clinical manifestations. Cases of rhabdomyolysis as the initial presentation of SLE are rare, and there are no reported cases of SLE presenting hyponatremia-associated rhabdomyolysis as the first manifestation. Herein, we report a case of SLE with lupus nephritis in a patient with acute hyponatremia-associated rhabdomyolysis.

PATIENT CONCERNS

A 44-year-old woman was admitted with complaints of altered consciousness, myalgia, and red-brownish urine that first appeared three days prior. Peripheral blood tests revealed elevated creatine kinase (19,013 IU/L) and myoglobin (5099 U/L) levels and severe hyponatremia (111 mEq/L) with no azotemia. Urinalysis showed nephritic sediments.

DIAGNOSIS

Whole-body bone scintigraphy showed increased uptake of radiotracer in the both upper and lower extremities. Serological evaluation revealed the presence of anti-nuclear (speckled pattern, 1:640), anti-double stranded DNA, and anti-Smith antibodies and absence of anti-Jo-1 antibody. A kidney biopsy demonstrated mesangial proliferative (class II) lupus nephritis.

INTERVENTIONS

Fluid therapy, including intravenous administration of 3% NaCl, was initiated. After three consecutive days of intravenous methylprednisolone (1 g/d), oral prednisolone (1 mg/kg/d), mycophenolate mofetil, and hydroxychloroquine were administered.

OUTCOMES

On day 28, the patient was discharged with marked resolution of SLE-associated symptoms and laboratory findings. Lupus reactivation was not present during the subsequent six-month follow-up.

LESSONS

Hyponatremia-associated rhabdomyolysis can be the first manifestation of SLE. Moreover, prompt fluid therapy and timely administration of immunosuppressive agents in SLE patients presenting with hyponatremia and rhabdomyolysis can significantly help alleviate disease activity and improve clinical outcomes.

Hyponatremia as an Inflammatory Marker of Lupus Activity Is a Fact or Fad: A Cross-Sectional Study

Purpose

Systemic lupus erythematosus (SLE) is an autoimmune disease with multi-organ involvement. Hyponatremia is common in inflammatory diseases such as meningitis. Also, it has been found to be strongly associated with the degree of inflammation. However, it has not been fully evaluated in lupus. This study aimed to assess the inflammatory impact of hyponatremia in SLE patients and investigate its relationship with SLE disease activity.

Patients and Methods

A total of 101 patients with SLE were enrolled in this study and divided into two groups according to Na level; a normo-natremic group and a hyponatremic group. Demographic and clinical data were collected. SLE activity was assessed by the systemic lupus erythematosus disease activity index (SLEDAI). The estimated sedimentation rate (ESR) as well as levels of C-reactive protein (CRP) and complements (C3 and C4) were measured.

Results

The majority of patients were females (98 subjects) (97%) with a mean age of 33±8 years. Out of 101 participants, 40 patients (39.6%) were hyponatremic with a mean Na level of 131.58±3.11 mmol/L. There was a statistically negative correlation between both ESR and SLEADI score and Na level (r=-0.436 and -0.436, respectively) with p=0.002, whereas Na level was positively correlated with complements, Cl, and albumin levels (r=0.653, 0.314, and 0.460, respectively) (p=0.000, 0.027, and >0.001, respectively). CRP was not correlated with Na level. ESR was independently correlated with hyponatremia at the 95% CI for Exp B (0.997-0.058) with a p-value of 0.048.

Conclusion

Hyponatremia could be used as an indicator of SLE activity. Also, it might be an easily and rapidly detected as well as a prognostically useful marker of inflammation.