A Societal Burden
Vision is central to most aspects of daily life. Multiple surveys report that Americans fear blindness more than most other disabilities, including losing a limb, deafness, using a wheelchair, and mental illness.
Age-related macular degeneration (AMD) is a leading cause of blindness around the world. It is a multifactorial disease with many risk factors. Nonmodifiable risk factors include age, genetic predisposition, and female sex. Modifiable risk factors such as smoking and the consumption of specific antioxidants and minerals are based on the prospective randomized Age-Related Eye Disease Study (AREDS) and AREDS2. Other dietary modifications and lifestyle habits may also affect the AMD risk profile.
Calcium and the Human Body
Calcium is the most abundant mineral in the body. It plays key physiologic roles in muscle function, nerve transmission, vascular tone, and hormone secretion. Even while participating in these important functions, the majority of calcium in the human body, approximately 99%, is stored in bones and teeth. It can be ingested from foods, medications (eg, antacids), or as a supplement. Approximately 43% of the US population and up to 70% of older women regularly consume dietary supplements containing calcium.
In the short term, inadequate calcium consumption typically produces no symptoms. Serum calcium is strictly regulated and typically does not vary with dietary fluctuation because bones serve as a reservoir. Bone itself undergoes continuous remodeling, a balance between resorption and deposition. In older adults, particularly among postmenopausal women and especially in the setting of inadequate calcium intake over the long term, bone resorption can exceed formation and potentially result in osteopenia.
Osteopenia, or low bone mass, affects approximately 34 million Americans and can lead to osteoporosis if left unmanaged. Osteoporosis affects more than 10 million American adults and is characterized by brittle and fragile bones. It is a common cause of fractures, including those involving the wrist, hip, vertebrae, pelvis, and ribs, and is responsible for an estimated 1.5 million fractures annually in the United States alone. Critically, supplementation with calcium and vitamin D is effective in reducing fractures in older adults.
Because of the key role of calcium in human health and skeletal integrity, the Institute of Medicine of the National Academies established a recommended dietary allowance (RDA), defined as the average daily intake that is sufficient to meet the nutrient requirements of nearly all (97%-98%) healthy individuals. All adults who are 19-50 years of age are recommended to consume 1000 mg of calcium daily; women older than 50 years of age and men older than 70 years are recommended to increase calcium intake to 1200 mg daily. These amounts can be achieved through a combination of diet and supplements.
AMD and Calcium Supplementation: Recent Controversy
Given the considerable prevalence of AMD and widespread calcium supplementation among Americans, a recent exploratory analysis of the relationship between AMD and calcium supplementation has garnered significant attention. Unfortunately, the published conclusion is misleading and has led to unfounded concern and inappropriate consideration of cessation of suitable calcium supplementation among at-risk Americans.
Kakigi and colleagues used the public National Health and Nutrition Examination Survey (NHANES) database from 2007-2008 to investigate the relationship between AMD and self-reported calcium supplementation. The Centers for Disease Control and Prevention (CDC) performs the NHANES annually to assess the health status of the US population. It typically includes an interview questionnaire and physical examination of approximately 5000 participants. In the 2007-2008 NHANES, fundus photographs were also obtained for participants who were 40 years of age or older and were graded for the presence or absence of AMD by trained professionals at the University of Wisconsin.
Of the 3191 participants who were 40 years of age or older, 248 (7.8%) were graded as having AMD: 220 with “early AMD” and 28 with “late AMD.” The authors did not define “early” and “late” AMD. Presumably, “late AMD” referred to advanced AMD, defined as either neovascularization or the development of geographic atrophy, but this was not clearly stated. Of these 248 participants with AMD, 146 adults (59%) reported taking calcium supplements. The authors then divided this relatively small sample size into quintiles based on the daily milligrams of calcium supplements consumed as self-reported by the participants. The prevalence of AMD was then assessed within each quintile. Each calcium supplement quintile had 23-44 total patients—small numbers to make substantial population-wide inferences.
As expected with such an epidemiologic study, there were significant differences between the populations with and without AMD. For example, participants with and without AMD were 67.2 and 55.8 years old, respectively, a difference of more than 11 years. Similarly, participants with AMD were significantly more likely to be smokers and reported no alcohol consumption. The authors adjusted for these and at least 14 confounders before reporting an increased prevalence of AMD among participants in the highest quintile of calcium supplement use (more than 800 mg daily) compared with those who did not report consuming calcium supplements (odds ratio of 1.85; associated 95% confidence interval, 1.25-2.75). Only 44 AMD patients reported daily calcium supplementation with more than 800 mg, representing 1.4% of the total population studied.
Many inconsistencies within the identified association between AMD and calcium supplementation must be considered. For example, no association between AMD and calcium supplement use was found among participants younger than 68 years of age. Rather, this relationship was only identified in participants older than 68 years of age. Furthermore, no dose-response relationship was identified between quintiles of calcium intake and AMD, a finding that one might anticipate if calcium served a legitimate role in AMD pathogenesis. The authors concluded that the association “suggests a threshold above which susceptible individuals are most vulnerable to the effects of calcium supplementation.” This is merely a hypothesis, one that may just as likely be correct as incorrect.
Even with sophisticated statistical analyses that include multivariate logistic regression modeling, application of their results remains fundamentally limited. Most important is that this is an observational analysis and therefore cannot demonstrate causation. Demonstrating that calcium supplements cause a particular outcome would require a longitudinal analysis in which new cases of AMD were identified among participants consuming varying amounts of calcium. This report described an interesting hypothesis-generating observation, which demands substantial further investigation before clinical guidelines are reconsidered.
Beyond this critical limitation, other major limitations are evident. First, NHANES is a questionnaire-based assessment for which participants recall the supplements they ingested within the immediately preceding 30 days. From these data, it is impossible to determine the duration of supplementation. The analysis did not differentiate between participants who consumed calcium supplements for less than 30 days and those who had been consuming calcium supplements for much longer.
Second, dietary calcium from sources other than supplements and antacids was not assessed. Therefore, total calcium intake is unknown and unable to be analyzed. This is particularly relevant in light of the recent analysis of the prospective observational Blue Mountain Eye Study (BMES) involving 2037 adults, with apparently discordant results. Gopinath and colleagues identified an increased risk of developing incident late AMD among participants with lower total dietary calcium intake over 15 years. When considering the relationship between calcium consumption and human diseases such as AMD, total calcium intake, including both dietary and supplemental sources, must be considered.
Finally, there may be limited longitudinal reliability of supplement consumption reporting in NHANES. For example, in NHANES I, 34% of participants who reported consuming daily supplements were no longer daily users at follow-up, and 25% of participants who reported no supplements were taking supplements at follow-up, suggesting that many participants change their supplement consumption behavior over time.
Summary and Consideration of Clinical Guidelines
Epidemiologic-based research is important for identifying and following health and disease trends among populations. However, as a science-based medical community, we must be careful to both highlight the limitations of research and refrain from overinterpreting results. Cross-sectional analyses such as the one reported by Kakigi and colleagues are excellent for generating hypotheses, but these hypotheses must be analyzed in other databases and, ideally, prospectively evaluated before causation can be seriously considered.
Therefore, clinical recommendations for calcium supplementation should not be changed on the basis of the data published by Kakigi and colleagues. Specifically, patients should not stop taking calcium supplements in an attempt to alter their risk for AMD development, especially given the exceptionally well-documented benefits of calcium supplementation in specific patient populations, including those at risk for osteoporosis. Rather, clinical advice to patients with AMD and those at risk for AMD should focus on well-documented strategies, including maintaining a nonsmoking lifestyle, optimal cardiovascular risk factor control, and consideration of AREDS-based supplementation.