Session Information
Session Type: ACR Concurrent Abstract Session
Session Time: 4:30PM-6:00PM
Background/Purpose: Histopathologic studies of cutaneous neonatal lupus erythematosus (cNLE) lesions usually show interface dermatitis. Hydroxychloroquine (HCQ) is an effective treatment for interface dermatitis seen in connective tissue diseases. It may also be effective for cNLE. Due to the transplacental passage of HCQ, fetuses of treated mothers are exposed to HCQ. The aim of this study was to asses if in utero exposure to HCQ would lower the risk of cNLE.
Methods: A multicenter case control study was performed. Inclusion criteria were: (1) infant born to a woman positive for anti-Ro ± anti-La antibodies and with a diagnosis of either systemic lupus erythematosus, Sjogren’s syndrome, dermatomyositis or rheumatoid arthritis, (2) infant cNLE status documented, and (3) documentation of maternal medications during pregnancy. Children with cardiac NLE were excluded from this study. Generalized estimating equation (GEE) was used to account for within-family correlation of data. A subgroup analysis was performed including only cases with cNLE onset ≤ 4 weeks of life, as neonates are not continued on HCQ after delivery therefore HCQ levels are expected to decline after birth.
Results: A total of 545 children (10 twin pairs) were included. Cases (N=112) and controls (N=433) differed with respect to maternal characteristics (Table 1). 169 (31%) infants were exposed to HCQ. Univariable GEE models showed that maternal diagnosis, anti-La antibody, maternal intake of HCQ and maternal intake of non-fluorinated steroids ± azathioprine were associated with cNLE (Table 2). In multivariable GEE, maternal intake of HCQ remained associated with a significant decrease in cNLE (Table 3). 41/112 infants with cNLE had onset of rash within 4 weeks after birth. When analyses were restricted to these early onset cNLE cases, results were similar to that of the entire study population (multivariable GEE, maternal intake of HCQ OR 0.2 (95% CI 0.1-0.7); p=0.009).
Conclusion: In this large multicenter study of 545 children born to women with anti-Ro ± anti-La antibodies and a connective tissue disease and/or RA, in utero HCQ exposure was associated with a decreased risk of cNLE.
| Table 1. Baseline characteristics | ||||
|
Patients (N) |
Cases (N=112) |
Controls (N=433) |
p value |
|
| Maternal characteristics1 |
535 |
|
|
|
| Age at child birth, years (median (IQR)) |
522 |
31 (29-35) |
32 (29-35) |
0.446 |
| Diagnosis, N (%) |
535 |
|
|
<0.001 |
| Sjogren’s syndrome |
|
31 (27.7) |
116 (27.4) |
|
| SLE |
|
52 (46.4) |
261 (61.7) |
|
| Dermatomyositis |
|
0 |
1 (0.2) |
|
| RA or juvenile idiopathic arthritis |
|
3 (2.7) |
15 (3.6) |
|
| Sjogren’s syndrome and SLE |
|
26 (23.2) |
29 (6.9) |
|
| Sjogren’s syndrome and RA |
|
0 |
1 (0.2) |
|
| Anti-Ro antibody positive, N (%) |
535 |
112 (100) |
418 (99) |
0.589 |
| Anti-La antibody positive, N (%) |
507 |
79 (72) |
191 (48) |
<0.001 |
| Medication intake, N (%) |
|
|
|
|
| Hydroxychloroquine |
535 |
19 (17) |
145 (34) |
<0.001 |
| Fluorinated steroids ± IVIG ± plasmapheresis |
535 |
7 (6) |
17 (4) |
0.310 |
| Non-fluorinated steroids ± azathioprine |
532 |
31 (28) |
185 (44) |
0.002 |
| Children characteristics |
545 |
|
|
|
| Gender, female:male, N |
543 |
65:47 |
207:224 |
0.059 |
| Born ≥ year 2000, N (%) |
545 |
64 (57) |
277 (64) |
0.183 |
| Age of onset of cNLE, weeks (median (IQR)) |
108 |
6 (3-10) |
N/A |
N/A |
1 N=535 due to 10 twin pregnancies
| Table 2. Variables associated with cNLE in univariable GEE | ||
| Maternal characteristics |
OR (95% CI) |
p value |
| Age at child birth |
1.0 (0.9-1.03) |
0.517 |
| SS diagnosis |
2.0 (1.3-3.1) |
0.002 |
| Anti-La antibody positive |
2.8 (1.7-4.4) |
<0.001 |
| Intake of HCQ |
0.4 (0.2-0.7) |
0.001 |
| Intake of fluorinated steroids ± IVIG ± plasmapheresis |
1.5 (0.6-3.8) |
0.387 |
| Intake of non-fluorinated steroids ± azathioprine |
0.5 (0.3-0.8) |
0.002 |
| Children characteristics |
|
|
| Female gender |
1.5 (1.0-2.2) |
0.054 |
| Born ≥ year 2000 |
1.4 (0.9-2.1) |
0.158 |
| Table 3. Variables associated with cNLE in multivariable GEE | ||
| Maternal characteristics |
OR (95% CI) |
p value |
| Anti-La antibody positive |
2.5 (1.6-4.1) |
<0.001 |
| Intake of HCQ |
0.4 (0.2-0.8) |
0.008 |
| Children characteristics |
|
|
| Female gender |
1.7 (1.1-2.7) |
0.012 |
To cite this abstract in AMA style:
Barsalou J, Costedoat-Chalumeau N, Berhanu A, Fors-Nieves C, Shah U, Brown P, Laskin C, Morel N, Levesque K, Buyon JP, Silverman E, Izmirly PM. Impact of in Utero Hydroxychloroquine Exposure on the Risk of Cutaneous Neonatal Lupus Erythematosus [abstract]. Arthritis Rheumatol. 2016; 68 (suppl 10). https://acrabstracts.org/abstract/impact-of-in-utero-hydroxychloroquine-exposure-on-the-risk-of-cutaneous-neonatal-lupus-erythematosus/. Accessed .« Back to 2016 ACR/ARHP Annual Meeting
ACR Meeting Abstracts - https://acrabstracts.org/abstract/impact-of-in-utero-hydroxychloroquine-exposure-on-the-risk-of-cutaneous-neonatal-lupus-erythematosus/