BACKGROUND AND AIM: Corticosteroid therapy is an effective way of treatment for many renal diseases, however, it is sometimes associated with infections. Our aim is to identify useful predictive markers of infection during steroid therapy. METHODS: We examined 121 patients (M/F = 71/50, mean age 43.8, range 15 - 82 years) who were treated with corticosteroids (IgA nephropathy in 51, minimal-change disease in 17, membranous nephropathy in 16 rapidly progressive glomerulonephritis (RPGN) in 13, lupus nephritis in 12 and other disorders in 12). Karnofsky's performance score (KPS) was employed to assess the physical functional status at the time of diagnosis. Infections were defined as conditions that required more than 1-week care, and those that caused the patient's death. RESULTS: Nineteen patients (15.7%) had infections during treatment. A logistic multivariate analysis showed significant correlations between infection and the use of immunosuppressive agents (relative risk RR = 7.7, p = 0.0265), ages of 52.9 years or more (RR = 13.5, p = 0.0026), initial number of lymphocytes (Lym) less than 1.250/microl (RR = 14.2, p = 0.0011), and KPS less than 77.4 (RR = 12.1, p = 0.0020). All correlations with infection were independent of all the other variables listed above. CONCLUSION: KPS, along with age, Lym and the use of immunosuppressive agents, are useful for the prediction of infectious complications during steroid therapy.
BACKGROUND AND AIM: Corticosteroid therapy is an effective way of treatment for many renal diseases, however, it is sometimes associated with infections. Our aim is to identify useful predictive markers of infection during steroid therapy. METHODS: We examined 121 patients (M/F = 71/50, mean age 43.8, range 15 - 82 years) who were treated with corticosteroids (IgA nephropathy in 51, minimal-change disease in 17, membranous nephropathy in 16 rapidly progressive glomerulonephritis (RPGN) in 13, lupus nephritis in 12 and other disorders in 12). Karnofsky's performance score (KPS) was employed to assess the physical functional status at the time of diagnosis. Infections were defined as conditions that required more than 1-week care, and those that caused the patient's death. RESULTS: Nineteen patients (15.7%) had infections during treatment. A logistic multivariate analysis showed significant correlations between infection and the use of immunosuppressive agents (relative risk RR = 7.7, p = 0.0265), ages of 52.9 years or more (RR = 13.5, p = 0.0026), initial number of lymphocytes (Lym) less than 1.250/microl (RR = 14.2, p = 0.0011), and KPS less than 77.4 (RR = 12.1, p = 0.0020). All correlations with infection were independent of all the other variables listed above. CONCLUSION: KPS, along with age, Lym and the use of immunosuppressive agents, are useful for the prediction of infectious complications during steroid therapy.