Abstract
Background: Folate deficiency is classically taught as a cause of anemia; with the advent of widespread dietary fortification, the prevalence of folate deficiency in the general population has decreased significantly. Recent studies have examined the yield of testing for folate deficiency and questioned its value, demonstrating negligible clinical yield across multiple care settings. However, patients with cancer are at increased high risk of folate deficiency. To our knowledge, no studies have specifically examined patterns and yield of folate testing among inpatients at a cancer institution. We sought to determine the yield of folate testing in this setting, and to characterize the patients found to have folate deficiency.
Methods: We used our institutional clinical database to identify all inpatient folate and red blood cell folate (RBCF) tests done at our institution in calendar-year 2017. We included testing done in our Urgent Care Center in patients who were subsequently admitted to the hospital. We defined folate deficiency as < 4.0 ng/mL according to our lab reference. Among patients with folate deficiency, we performed manual chart review to evaluate the presence of known risk factors for folate deficiency: alcohol abuse, gastrointestinal malabsorption, pancreatic insufficiency, culprit medications, end-stage renal disease requiring hemodialysis, malnutrition (defined as BMI <18.5), and hemolytic anemia.
Results: A total of 1013 serum folate level tests were performed in 960 patients; one patient underwent RBCF testing alone. A majority of tests were done on medical rather than surgical patients (85% vs 13%). Overall, 69 (6.8%) of patients were found to have folate deficiency. Almost all patients with folate deficiency were anemic. Surprisingly, macrocytosis was infrequent in patients with folate deficiency in cancer; over 90% of patients had normal or even low MCV. Among 69 patients with folate deficiency, less than half the patients (N=28, 41%) had an identifiable risk factor for folate deficiency, most commonly medications. Identified malabsorption/ malnutrition, hemolysis, and hemodialysis collectively accounted for 7.2%, and no case was attributed to alcoholism. The degree of anemia was not significantly different between the patients with folate deficient and those with adequate folate levels. The mean MCV values were normal in both the folate deficient and folate replete patients. Co-testing of vitamin B12 levels was performed in 859 (85%) of testing instances; 23 (2.7%) had isolated B12 deficiency and 3 (0.3%) had concurrent folate and B12 deficiency.
Conclusions: In our cohort of hospitalized cancer patients, folate testing identified deficiency in 6.8%, notably higher than in other studies of general populations. A higher rate of folate deficiency may have been present in this study due to the comorbidities associated with advanced malignancies and their treatments. Surprisingly, folate deficiency did not correlate with macrocytosis, which is characteristic of the folate deficiency-associated anemia in non-cancer patients. This may reflect more acute development of deficiency, or the presence of multiple factors affecting red cell production, thus "masking" the macrocytosis. Neither the degree of anemia nor the MCV provided guidance as to the presence of folate deficiency and a minority of deficient patients had identifiable risk factors. Folate deficiency, while not common in hospitalized cancer patients, needs to be considered in the presence of anemia, even in the absence of macrocytosis.
Soff:Amgen: Research Funding; Janssen: Research Funding.
