Association between use of low-dose aspirin on hemoglobin levels and serum iron homeostasis in patients with acute coronary syndrome or ischemic stroke

Abstract

Background: 

Association between low-dose aspirin (LDA) and possible covert effects on hemoglobin (without overt bleeding) has been inconsistent. The current study was planned to assess the effects of LDA on hemoglobin level and serum iron status over a period of 6 months in nonanemic patients with acute coronary syndrome (ACS) or ischemic stroke.

Methods: 

This was a hospital-based open cohort prospective study conducted on newly diagnosed patients with ACS or ischemic stroke discharged on secondary prophylaxis of LDA (<162 mg/day) The study period was 6 months of patient recruitment and 6 months of follow-up.

Results: 

Out of originally recruited 81 patients, 52 were followed up for the entire 6 months. The mean age was 61.8 ± 8 years. In 43 (82.6%) patients, the hemoglobin level had a fall at 6 months from the baseline level. Male patients had a significant fall in hemoglobin levels while as females had an insignificant change. Anemia was detected in 20 (38%) cases at the end of 6 months. The incidence of anemia was 0.38 cases/180 days. A fall in serum ferritin was observed in 30 patients and a rise was observed in 22 patients. At the end of 6 months, serum ferritin fell to <20 μg/L in five patients, out of which two patients had serum ferritin <15 μg/L.

Conclusion: 

The study provided insight about the effect of LDA on the hemoglobin level and serum iron profile among the patients who were started on LDA for the secondary prophylaxis.

Introduction

Low-dose aspirin (LDA) is taken by an increasing number of people for primary and secondary prophylaxis of atherothrombotic diseases, dementia, and some malignancies. These conditions are common in elderly and changes in demographics mean that there is potential for many older people to take LDA.[¹] Aspirin when used as low dose (<162 mg) has an antiplatelet action.[²] The most striking observation is that there is so little relevant information on possible covert effects on hemoglobin with such a commonly prescribed drug such as aspirin despite its well-known propensity to cause significant overt gastrointestinal bleeds. The relevant available data shows no consistent association between LDA and anemia. While some studies indicated that hemoglobin levels are reduced, others show no change, or even suggest a higher hemoglobin level.[¹] Due to the importance of anemia and the uncertainty as to whether taking LDA has a clinically significant effect on hemoglobin (other than in the context of overt bleeding) this study was conducted to observe any evidence of an association between use of LDA on hemoglobin levels and serum iron status.

Methods

This was a hospital-based open cohort prospective analytical study which was conducted in a tertiary care referral hospital in Himachal Pradesh, India. The study period was of 1 year which included 6 months of patient recruitment and 6 months of follow-up. Recruitment of patients was done from January, 2019 to June, 2019 using an open cohort design. The inclusion criteria were newly diagnosed patients with acute coronary syndrome (ACS) and ischemic stroke. Patients were discharged on LDA (<162 mg/day). In addition to LDA, patients of ACS received clopidogrel 75 mg once a day. The exclusion criteria were (a) patients with anemia, (b) presenting with active gastrointestinal bleed or bleed from any other site, (c) history of peptic ulcer disease, (d) patient already on LDA, (e) patient receiving nonsteroidal anti-inflammatory drugs, and (f) did not give informed consent. Baseline complete hemogram and serum iron profile including serum ferritin, serum iron, total iron binding capacity (TIBC) and percent transferrin saturation was done in all the recruited patients. Follow-ups were done at regular intervals during hospital visits or through a telephonic conversation with the patient and/or caregiver for the next 6 months. During the follow-up, any adverse events, hospitalization, death was recorded. Complete hemogram and iron status parameters serum ferritin, serum iron, TIBC, and percentage transferrin saturation were again assessed at the end of 6 months. The data were entered into MS office and analyzed using SPSS version 21 (IBM corporation, Armonk, New York, USA). Quantitative variables were expressed as mean with standard deviation and categorical variables as frequencies and percentage. For analysis of continuous variables Student's t-test was applied and P < 0.05 was considered statistically significant. The study was approved by institutional ethics committee vide no IEC/2019 dated January 10, 2019.

Definitions

• LDA: Aspirin <162 mg/day[²]
• Anemia: Hemoglobin <12 mg/dl in female and <13 mg/dl in male patients[³]
• Negative iron balance: Serum ferritin <20 ug/L[⁴]
• Iron deficiency erythropoiesis: Serum ferritin <15 ug/L[⁴]
• Iron deficiency anemia: WHO definition of hemoglobin <12 mg/dl in female and <13 mg/dl in male patients and serum ferritin <15 ug/L.[³⁴]

Results

A total of 81 patients documented to have ACS or ischemic stroke were recruited in this study. During the study, 22 patients were untraceable hence lost in the follow-up. Five patients died, out of which three were of ACS and two were of stroke. One patient each developed bleeding hemorrhoids and hematemesis hence were excluded. Thus, a total of 29 (35.8%) patients were excluded from the original recruited cohort in this study. Fifty-two patients were followed up for the entire 6 months. Out of 52 patients, 41 (78.8%) patients were of ACS and 11 (21.2%) were of ischemic stroke. On gender distribution, 44 (84.6%) were male and 8 (15.4%) were females. The male-to-female ratio was 5.5:1. Among male patients, 38 were of ACS and 6 were of stroke, and among females, 3 were of ACS and 5 were of stroke. The mean age was 61.8 ± 8 years. The number of patients in the age group of 40–50 years was 5 (9.6%), 51–60 years was 18 (34.6%), 61–70 years was 23 (44.2%), 71–80 years was 6 (11.5%). [Table 1] shows hemogram and iron profile at admission and at 6 months among the study subjects. In 43 (82.6%) patients, the hemoglobin level had a fall at 6 months from the baseline level at admission. The striking observation was that anemia was detected in 20 (38%) cases at the end of 6 months. Nineteen cases were males and one was female. Of 20 cases 16 were of ACS and four cases were of acute ischemic stroke. The incidence of anemia was 0.38 cases per 180 days. A fall in hemoglobin level of ≤1 g/dl was observed in 26 (60.4%) patients; 1.1 to ≤2 g/dl in 8 (18.6%) patients; 2.1 to ≤3 g/dl in 4 (9.3%) patients; and ≥3.1 g/dl in 5 (11.6%) patients. The hemoglobin level increased in 5 (9.6%) patients and in 4 (7.6%) patients remained unchanged from the baseline at 6 months' follow-up. Baseline serum ferritin was ≤15 μg/L in two patients. Both were nonanemic thus reflecting iron deficient erythropoiesis. A fall in serum ferritin was observed in 30 patients and a rise was observed in 22 patients. At the end of 6 months, serum ferritin fell to <20 μg/L (developed negative iron balance) in five patients, out of which two patients had serum ferritin <15 μg/L (iron deficiency erythropoiesis -both had a fall in hemoglobin levels but did not develop anemia). Table 2 shows hemogram and iron profile at admission and at 6 months in male and female patients, respectively. Table 3 shows hemogram and iron profile at admission and at 6 months in patients with ACS and ischemic stroke.

Table 1: Hemogram and iron profile at admission and at 6 months among patients

Table 2: Hemogram and iron profile at admission and at 6 months in male and female patients

Table 3: Hemogram and iron profile at admission and at 6 months in patients with acute coronary syndrome and stroke

Discussion

Aspirin has various adverse effects such as nausea, vomiting, epigastric distress, hypersensitivity, salicylism, and occult blood loss in stool. The most important adverse effect is mucosal damage and peptic ulceration. About 1 in 100 patients taking aspirin over a 28-month period experience gastrointestinal hemorrhage.[⁵] The impact of cumulative iron losses from occult bleeding associated with daily use of aspirin has been compared with typical cumulative menstrual iron losses.[⁶] In a systematic review, the weighted mean fecal blood loss with LDA (arbitrarily set at ≤325 mg/day) was 1.6 ml/day.[⁷] Sustained iron depletion due to long term LDA use has been suggested as one of the mechanisms of cardioprotective effect.[⁶] Long-term daily LDA intake has shown benefits on cancer mortality and chronic iron loss has been suggested as a plausible antitumor mechanism.[⁸⁹] In this study, we observed a significant fall in the mean hemoglobin level and mean packed cell volume (PCV) among patients at 6 months' follow-up. A subgroup analysis of the results revealed a significant fall in the mean hemoglobin level and mean PCV among male patients in comparison to female patients. Patients on single and dual antiplatelets were reviewed separately and in patients of ACS who were on dual antiplatelets had a significant fall in mean hemoglobin level and mean PCV. A fall in hemoglobin was observed in 43 (82.6%) patients and 20 (38%) patients developed anemia. The adverse effects of LDA (100 mg/day) have been studied in a double-blind randomized, placebo-controlled trial for primary prevention among healthy, elderly subjects above 70 years of age and conducted over a period of 12 months. It was observed that aspirin treated persons had a decrease in mean hemoglobin levels of 0.33 g/dL compared with 0. 11 g/dL in the placebo group (P < 0.05). A decline of 1 g/dL or greater was observed among 17.4% of patients ingesting aspirin.[¹⁰] In a retrospective study on 80 patients, who received LDA for secondary prevention, a significant loss of mean hemoglobin by 0.472 g/dl in men and an insignificant change in hemoglobin level in women was observed.[¹¹] The explanation for the sex related difference has been that the benefit of aspirin in reducing stroke or death was restricted to men as the antithrombotic effects of aspirin differs between sex and apparent predisposition to occult blood loss among men.[¹²] On the contrary, anemia was found to be 42% less common among aspirin users in a cross-sectional analysis on the relationship between LDA and presence of anemia among a cohort of 464 patients aged 77 years.[¹³] The explanation for these unexpected results in the study was that the aspirin users were less likely to have peptic disease, physicians withheld aspirin from anemic subjects even without peptic disease and better follow up thus more likely to receive therapy for anemia. In fact, LDA has been postulated to diminish cytokine release and its inflammatory consequences and may on balance prevent anemia even while promoting bleeding to a small degree; dampening in the suppression of erythropoietic hemoglobin synthesis due to chronic disease. However, the data were cross-sectional and not longitudinal.[¹³] In the present study, we analyzed the age difference among males and females as a possible cause for the significant fall in the mean hemoglobin level and mean PCV among male patients. However, the mean age for males and females is similar and is 61.9 ± 8.4 years and 61.9 ± 4.1 years, respectively. Iron status was measured by serum iron, serum ferritin, TIBC, and transferrin saturation. A significant fall in mean serum iron and mean percent transferrin saturation was observed among patients at 6 months. Sex-related subgroup analysis revealed a fall in mean serum iron, ferritin, percentage transferrin saturation, and rise in TIBC in males. The fall in mean percentage transferrin saturation was significant. The difference in mean serum iron, ferritin, and TIBC among male was not statistically significant. There was fall in mean serum iron, ferritin, percentage transferrin saturation, and rise in TIBC among female patients but the difference was not statistically significant. Patients on single and dual antiplatelets were reviewed separately and in patients of ACS who were on dual antiplatelets had an insignificant fall in mean serum iron, ferritin and rise in TIBC. The fall observed in mean percent transferrin saturation among patients with ACS was statistically significant. Iron studies did not correlate with the change in hemoglobin levels in this study. Even though hemoglobin concentration is the last iron index to change in uncomplicated iron deficiency, and thus it may not provide information about early stages of iron storage depletion, which instead is reflected by decreased serum ferritin concentration. However, a major drawback in using serum ferritin for estimating body iron stores is that ferritin synthesis and secretion by hepatic cells can be increased by inflammatory cytokines which could result in increased blood serum ferritin concentrations that do not reflect actual increases in iron stores.[¹⁴] Long-term aspirin use has been shown to be associated with approximately 20% lower serum ferritin concentrations.[¹⁵] Observational studies suggest the magnitude of aspirin mediated iron loss. In the elderly participants of the Framingham Heart Study, it was found that mean serum ferritin was lower among categories of regular aspirin users by 21%–50%.[¹⁴] In Danish DAN-MONICA 10 survey (men aged 40–70 year), aspirin (1–3 g/d) use was associated with serum ferritin values 20% lower than those of nonusers.[¹⁶] In addition, 6 months' follow-up may not have manifested all the biochemical parameters of iron deficiency which would have been statistically significant.

Conclusion

The study leads us to a conclusion that there is an association between LDA and a significant fall in hemoglobin levels among male patients in the absence of overt gastrointestinal bleed. The clinical significance of these findings is uncertain and the observation requires future research, keeping in view the burden of atherosclerotic cardiovascular diseases and use of LDA as primary and secondary prophylaxis.

Ethical approval statement

The study was approved by the Institutional Ethics Committee vide number IEC/2019 dated 10.1.2019.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.