Prevalence of Celiac Disease in Patients With Iron Deficiency Anemia—A Systematic Review With Meta-analysis

• Published in: Gastroenterology (New York, N.Y. 1943) Vol. 155; no. 2; pp. 374 - 382.e1
• Main Authors: Mahadev, Srihari, Laszkowska, Monika, Sundström, Johan, Björkholm, Magnus, Lebwohl, Benjamin, Green, Peter H.R., Ludvigsson, Jonas F.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01-08-2018
• Subjects: Age Factors; Anemia, Iron-Deficiency - epidemiology; Anemia, Iron-Deficiency - etiology; Anemia, Iron-Deficiency - pathology; Biopsy; Celiac; Celiac Disease - complications; Celiac Disease - epidemiology; Celiac Disease - pathology; Coeliac; Humans; Intestine, Small - pathology; Iron Deficiency; Medicin och hälsovetenskap; Meta-Analysis; Prevalence; Sex Factors; CD; Meta-Analysis; GFD; Coeliac; CI; Celiac; IDA; Iron Deficiency
• ISSN: 0016-5085; 1528-0012; 1528-0012
• DOI: 10.1053/j.gastro.2018.04.016

Anemia is common in patients with celiac disease (CD) and a frequent mode of presentation. Guidelines recommend screening patients with iron-deficiency anemia (IDA) for CD. However, the reported prevalence of CD in patients with IDA varies. We performed a systematic review to determine the prevalence of biopsy-verified CD in patients with IDA.
We performed a systematic review of articles published in PubMed Medline or EMBASE through July 2017 for the term “celiac disease” combined with “anemia” or “iron deficiency.” We used fixed-effects inverse variance-weighted models to measure the pooled prevalence of CD. Meta-regression was used to assess subgroup heterogeneity.
We identified 18 studies composed of 2998 patients with IDA for inclusion in our analysis. Studies originated from the United Kingdom, United States, Italy, Turkey, Iran, and Israel. The crude unweighted prevalence of CD was 4.8% (n = 143). Using a weighted pooled analysis, we found a prevalence of biopsy-confirmed CD of 3.2% (95% confidence interval = 2.6-3.9) in patients with IDA. However, heterogeneity was high (I2 = 67.7%). The prevalence of CD was not significantly higher in studies with a mean participant age older or younger than 18 years or in studies with a mixed-sex vs female-predominant (≥60%) population. On meta-regression, year of publication, female proportion, age at CD testing, and prevalence in the general population were not associated with the prevalence of CD in patients with IDA. In the 8 studies fulfilling all our quality criteria, the pooled prevalence of CD was 5.5% (95% confidence interval = 4.1–6.9).
In a systematic review and meta-analysis, we found that approximately 1 in 31 patients with IDA have histologic evidence of CD. This prevalence value justifies the practice of testing patients with IDA for CD.