Simplified lupus infection prediction score more accurate, feasible than previous version
Click Here to Manage Email Alerts
Key takeaways:
- SLESIS-R predicts severe infections in lupus more accurately than SLESIS.
- Its simpler design could make SLESIS-R valuable in daily clinical practice, observational studies and clinical trials.
The SLE Severe Infection Score-Revised, an updated score for predicting severe infections in systemic lupus erythematosus, is more accurate and feasible than the prior version, according to data published in Lupus Science & Medicine.
“Properly estimating the risk of infection in patients with SLE is paramount if we are to balance immunosuppression and implement preventive measures,” Inigo Rua-Figueroa, MD, of the Hospital Universitario Gran Canaria Doctor Negrin, in Las Palmas de Gran Canaria, Spain, and colleagues wrote. “No evidence-based, widely validated, and suitable score for predicting severe infection in patients with SLE has been developed for use in daily clinical practice.”
Rua-Figueroa and colleagues previously developed a prediction tool called the SLE Severe Infection Score (SLESIS) using retrospective, cross-sectional data from the Spanish Rheumatology Society Systemic Lupus Erythematosus Registry. However, the performance of SLESIS was “only moderate,” they wrote. The team aimed to improve the score by, among other means, adding new markers based on higher quality data from the prospective phase of the registry. The current multicenter analysis additionally sought to validate the updated measure, which the researchers called the SLE Severe Infection Score-Revised (SLESIS-R).
Their analysis included 1,459 patients (mean age, 49 years) who completed two study visits or had infections or died during the study period. A multivariable logistic model accounted for variables already part of the SLESIS score, plus others identified through literature review. The model’s performance was gauged using C-statistic and area under the receiver operating curve.
According to the researchers, the adjusted multivariate model showed that severe infection in the following year could be predicted based on four variables. They included age 60 years or older, previous SLE-related hospitalization, previous serious infection, and having received glucocorticoids 30 mg per day or greater. These variables were used to build a score with total values ranging from zero to 17.
The area under receiving operator curve for the updated measure was found to be 0.861 (95% CI, 0.777-0.946), comparing favorably with the original SLESIS result of 0.79 (95% CI, 0.73-0.85). The chosen cutoff score was six or greater, which demonstrated an accuracy of 85.9% and a positive likelihood ratio of 5.48.
The feasibility of SLESIS-R also improved over SLESIS, as it is simplified due to the exclusion of Katz severity index, according to the researchers. Internal validation of the model was conducted using a 100-sample bootstrapping procedure, which yielded “appropriate discrimination parameters” and a C statistic of 0.81, they wrote.
“Because of its simplicity and the fact that it is based on clinical parameters and not laboratory results, SLESIS-R could become a useful instrument for predicting infection in both daily clinical practice and observational studies and even in clinical trials in Caucasians,” Rua-Figueroa and colleagues wrote.
“In fact, the use of numerical probabilities is to be preferred not only for decision-making but also in teaching materials and in communication between physicians,” they added. “We think that our score improves prediction of the risk of infection, facilitating an informed decision-making process and supporting more careful implementation of preventive measures.”
Perspective
Back to Top
David A. McLain, MD, MACR, FACP, FRCP
Infection is a significant concern in systemic lupus erythematosus. Problems in both the innate and adaptive components of patients’ immune response compromise the immune system, making them more susceptible to infection.
A total of 26 immune-related defects have been identified in the SLE patient that can handicap the body’s response to infection. Furthermore, the treatment of lupus almost always involves the use of immunosuppressive drugs. The combination of the immunocompromised state and the medications used to treat SLE increases susceptibility to acute and chronic infections, resulting in higher morbidity and mortality.
Infections are responsible for about a third of hospitalizations and deaths in SLE. Bacterial infections are prevalent in patients with SLE, constituting two-thirds of all infections. More than 60% of reported cases involve infections in the respiratory and urinary tracts, as well as the skin, which are the most frequent infection sites. Herpes zoster is the most common viral infection among patients with SLE, with an incidence that is two to three times higher than that of the general population.
In their own cohort, the present authors found that glucocorticoid doses of 30 mg increased the RR for infection to 7.47, while cyclophosphamide use demonstrated an RR of 12.38, mycophenolate registered an RR of 3.04, and rituximab had an RR of 1.32. On the other hand, antimalarials resulted in an RR of 0.64, while methotrexate or azathioprine demonstrated an RR of 0.79.
Being able to predict infection risk in patients with SLE is very beneficial to the clinician. Prior attempts to do this involved using decreased serum albumin, increased serum creatinine and prednisolone ≥60 mg/day, but was for inpatients only. Another model incorporated the SLEDAI >10, lymphocyte count <0.8×109/L and serum creatinine >1.18 mg/dl. This latter model has yet to be validated.
The present risk predictor score (SLESIS- R) has been validated, is easy to use, and gives a numerical result that is sensitive and specific.
Collapse
Read more about
Click Here to Manage Email Alerts