Table 3

Univariate and multivariate logistic regression analysis of potential risk factors during 12-month follow up in pediatric-onset systemic lupus erythematosus (pSLE)*

Univariate analysis

Multivariate analysis

Variable

AMS < 3.7

AMS > 3.7

OR (95% CI)

P-value

OR (95% CI)

P-value

Osteopontin high

8

3

2.1 (0.39, 10.8)

0.4

NGAL high

7

2

0.8 (0.14, 4.8)

0.9

Cumulative prednisone exposure

12

3

0.7 (0.15, 3.2)

0.7

Disease duration > 2 years

16

4

0.6 (0.15, 3.2)

0.7

Renal involvement

22

7

1.4 (0.3, 8.3)

0.9

Non-Caucasian background

25

10

5.4 (0.28, 104)

0.3

Male gender

8

2

0.8 (0.04, 3.1)

0.9

ΔSDI = 0

ΔSDI > 0

Osteopontin high

6

4

9 (1.5, 60)

0.015

7.5 (2.9, 20)

0.03

NGAL low

8

2

1.3 (0.21, 8)

0.9

1.16 (0.14, 9.3)

0.9

Cumulative prednisone exposure

11

4

2.8 (0.5, 14.7)

0.4

Disease duration > 2 years

17

2

0.4 (0.08, 2.4)

0.4

Renal involvement

24

5

2.1 (0.2, 20.2)

0.9

Non-Caucasian background

28

7

3.4 (0.2, 68)

0.6

Male gender

10

0

0.6 (0.008, 3)

0.16

*Except where indicated otherwise, values reflect the number (%) of patients. Odd ratios (ORs) were determined for each variable based on a yes/no determination in patients with pSLE (n = 42). Potential risk factors identified by univariate analysis (P ≤ 0.1) were included in the multivariate models. Total number of SLE patients with adjusted mean SLE disease activity index (AMS) in the top quartile (AMS > 3.7) is 10; total number of SLE patients with change in SDI (ΔSDI) greater than 0 is 7. SDI, Systemic Lupus International Collaborating Clinics/American College of Rheumatology (SLICC/ACR) damage index; NGAL, neutrophil gelatinase-associated lipocalin.

Rullo et al. Arthritis Research & Therapy 2013 15:R18   doi:10.1186/ar4150