To the Editor: Anemia is a common condition in older people.1 It appears to have multiple etiologies, including nutritional deficiencies and chronic diseases.1,2 Nevertheless, the cause of approximately one-third of cases remain unexplained.1 Anemia is diagnosed in 23% to 60% of patients with hypothyroidism.3 In most cases, this anemia is reversible; a significant increase in hemoglobin levels after levothyroxine treatment has been observed in patients with hypothyroidism.3 Although, anemia is a common finding in adult and adolescent patients with hypothyroidism, there have been no systematic studies of the association between anemia and hypothyroidism in older adults, so no conclusions can be drawn about the indication of thyroid function screening in older people with anemia. This is a crucial point to take into consideration; aging is associated with a higher prevalence of hypothyroidism, with frequencies ranging from 0.5% to 5% for overt hypothyroidism and 5% to 20% for subclinical hypothyroidism in women aged 60 to 65 and older.4,5 In addition, the suspicion of hypothyroidism based on clinical symptoms is difficult in older subjects. Symptoms and signs of hypothyroidism in older adults are often insidious, ambiguous, and masked by other concomitant diseases, and many similarities between hypothyroidism and the aging process can be observed: bradycardia, hypertension, hypercholesterolemia, weight gain, cold intolerance, constipation, muscle weakness, lethargy, cognitive dysfunction, depression, impotence, and dry skin. The occurrence of anemia, defined according to the World Health Organization6 criteria, was evaluated in 316 patients aged 65 and older (mean age±SD 81.5±8.2, M/F: 114/202) in the Division of General Internal Medicine and Geriatrics, San Luca Hospital, between October 2007 and March 2009, and the prevalence of hypothyroidism in patients with and without anemia was compared. The effect of levothyroxine therapy on hemoglobin levels was also studied in a subgroup of subjects with hypothyroidism with mild anemia. All included participants underwent a complete medical interview and physical examination, and venous fasting blood samples were drawn for laboratory analyses (complete blood count, thyroid-stimulating hormone (TSH), complete biochemical profile). In all subjects with abnormal TSH, levels of free triiodothyronine (FT3) and free thyroxine (FT4) were determined. Patients who had received any red blood cell transfusion, erythropoiesis-stimulating agents, or iron supplements in the 4 months before study entry were excluded. Anemia was discovered in 155 of 316 (49%) enrolled patients (median hemoglobin 10.7 g/dL). Chronic disease (37.4%) and nutrient deficiencies (27.8%) were the most common causes of anemia. Approximately 14.8% of patients with anemia had other specified causes, and 5.2% had myelodysplastic syndrome. The etiology of anemia was unknown for 14.8% of subjects. The prevalence of hypothyroidism was significantly higher (chi-square P=.01) in patients with anemia (n=31, 20%) than in those without (n=16, 9.9%). Hypothyroidism was overt in eight and two subjects with and without anemia, respectively. TSH values were higher (P=.003) in participants with anemia (median 2.3 mU/L) than in those without (median 1.6 mU/L), whereas no significant differences in FT3 and FT4 values were reported between the groups. In subjects with hypothyroidism and anemia, TSH and hemoglobin levels were inversely correlated (correlation coefficient=−0.375, P=.04). In 21 subjects with hypothyroidism and anemia, there were no other potential causes of anemia, underlining the importance of TSH levels to avoid an erroneous diagnosis of unexplained anemia. Nine subjects with mild anemia and hypothyroidism were treated exclusively with levothyroxine. Exclusion criteria for the selection of these patients were hemoglobin level less than 10 g/dL, multifactorial anemia or anemia due to other reasons, cardiac ischemia, unstable angina pectoris, arrhythmias, uncontrolled hypertension, and diabetes mellitus. A significant decrease in TSH (Figure 1A) and increase in hemoglobin (Figure 1B) levels was reached with levothyroxine therapy. In all subjects, TSH levels normalized during treatment, and a median increase in hemoglobin of about 1.6 g/dL was observed at the time of TSH normalization. In addition, an inverse correlation between variations of TSH and hemoglobin levels during levothyroxine therapy was observed (correlation coefficient=−0.63, P=.07). These observations are important considering that, in older people, anemia appeared to be an independent predictor of mortality and is associated with greater morbidity, hospitalization, and disability and poorer physical performance, muscle strength, and cognitive function.7–9 These detrimental effects are observed not only in older individuals with severely low levels of hemoglobin but also in those with mild anemia.10 Thyroid-stimulating hormone (A) and hemoglobin (B) levels before and after treatment with levothyroxine. In conclusion, the high prevalence of hypothyroidism in people with anemia and the possibility of improving hematological parameters through levothyroxine treatment suggest that older hospitalized subjects with anemia should be screened for hypothyroidism according to TSH level. Conflict of Interest: The editor in chief has reviewed the conflict of interest checklist provided by the authors and has determined that the authors have no financial or any other kind of personal conflicts with this paper. Author Contributions: Giovanni Vitale was the principal investigator. Giovanni Vitale and Daniela Mari designed and coordinated the research. Letizia Maria Fatti, Silvia Prolo, and Andrea Girola recruited the patients. Michele Caraglia, Monica Marra, and Alberto Abbruzzese partially performed the laboratory work. Giovanni Vitale and Giancarla Gerli performed statistical analysis. Giovanni Vitale, Michele Caraglia, Giancarla Gerli, and Daniela Mari interpreted the data and wrote the letter. Sponsor's Role: No sponsor.
No takes yet. Share an insight, caveat, or question.
Vitale et al. (2010) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: