Background: Micronutrients are essential food components that are not a major energy source but are required in small quantities to support normal health. Vitamin A is well known for its role in vision; however, it is also involved in immune function, growth, and development. Deficiency of vitamin A manifests as xeropthalmia in children and night blindness. Chronic A deficiency is associated with respiratory diseases like pneumonia and increased risk of anemia (from iron deficiency) and death. Vitamin A deficiency is considered rare in the USA but is reported commonly in developing countries. Limited data is available in the USA about the prevalence of Vitamin A level testing and deficiency, especially in hospitalized patients. The purpose of this study is to evaluate the trends of vitamin A levels at Grady Memorial Hospital, a safety-net facility in Atlanta, Georgia that serves a large metropolitan area.
Methods: We performed a cross-sectional observational study of all admitted adult patients (age >18 years) whose serum vitamin A levels were tested at Grady hospital from January 2013 to November 2024 to determine the prevalence of abnormal (high or low) levels. Slicer Dicer was used to obtain patient data from the EPIC electronic medical record. Serum testing of vitamin A (< 38 mcg/dL) indicates deficiency. We followed up with a pilot study to look in greater depth at abnormal samples from 1/1/2023 to 12/31/2023 to examine the proportion that were low and the demographic characteristics of those with deficiencies. Microsoft Excel was used to perform descriptive statistics.
Results: Serum vitamin A levels were ordered 705 times between 2013–2024; 524 samples (74.3%) were labeled abnormal. In the pilot study from 1/1/2023 to 12/31/2023, among all 135 patients tested for Vitamin A, 107 (79.2%) came back “abnormal,” and all but 2 samples tested as deficiently low (< 38 mcg/dL). Fifty-nine samples (55%) were ≤20 mcg/dL, reflecting moderate deficiency, and 22 samples (20.6%) were ≤10 mcg/dL, reflecting severe deficiency. The median age of patients was 55.5 years. Sixty of the abnormal low samples (56%) were in men and 47 (44%) were in women. Eighty-six patients (80.3%) self-reported Black or African American race while 11 (10.2%) identified as White or Caucasian. One quarter (n=27, 25.2%) of patients had a BMI >30; 39 patients (36.4%) had a normal BMI (between 18.5 and 25); and 18 (16.8%) patients had BMI < 18.5. At the end of the study period, 93 patients (87%) were alive and 14 (13%) were deceased. Eight patients had repeat Vitamin A levels tested, but only the first value was included in the study.
Conclusions: To our knowledge, this is one of the largest data samples of vitamin A testing in admitted patients. Three in four samples tested for serum vitamin A level were lower than the minimum cut-off value. Though vitamin A deficiency in hospitalized patients may not reflect vitamin A deficiency in the general population, the high prevalence of vitamin A deficiency we observed is alarming. Also, with three in four patients tested for vitamin A showing deficiency, it is important to investigate the risk factors and specific patient populations that are at risk for developing vitamin A deficiency and subsequent clinical repercussions.