Don’t Let Your Patients Toss Them
The therapeutic effects of vitamin D and calcium supplementation have recently been called into question, largely as a result of the publicity surrounding 2 clinical trials that examined the benefit of these nutrients in reducing bone fractures and preventing colon cancer (N Engl J Med. 2006;354:669-683; 684-696). The largely negative media blitz that followed these publications has created a dilemma for physicians, who must now explain the results to patients who may be on the brink of tossing their supplements out the window. Several experts now question these results and offer advice on what you could tell your patients.
The calcium plus vitamin D trial, part of the Women’s Heath Initiative (WHI), was designed primarily to study their effects on hip fracture; secondary end points were effects on other ?fractures and on colorectal cancer. This 7-year, randomized, ?placebo-controlled study in?cluded 36,282 postmenopausal women (aged 50-79). Half of them received 1000 mg of calcium carbonate combined with 400 IU of vitamin D daily, and the other Half Received a Placebo. a Subset of Women had Regular Bone Density Scans. at the end of the Study, Three Quarters of the Participants Were Still Taking all the Pills.
“Women who adhered to study medication...had a significant 29% reduction in risk of hip fracture, and women aged 60 and older had a significant 21% reduction in risk.”
JoAnn E. Manson, MD, DrPh
Women who took the supplements had a 1% higher hip bone density than women who took placebo. During the trial, 374 women had hip fractures. A nonsignificant 12% reduction in hip fracture was found in the treatment group; hip fracture incidence was reduced by a significant 29% in women who were fully compliant and by 21% in women >60 years old. No significant differences were seen in spine or total fracture rate compared with the placebo group.
Major Study Design Flaws, Misinterpretation of Results
According to Robert P. Heaney, MD, John A. Creighton University Professor, Creighton University, Omaha, Neb, the observed fracture rate in the study was about half of what had been predicted. However, although the dose of supplemental calcium was probably correct, half of the women taking calcium were also taking estrogen, and half were obese (body mass index >30 kg/m2). “Whatever other health effect estrogen and obesity may have, both are osteoprotective and reduce the risk of fracture,” Dr Heaney told IMWR.
Another important factor, and one that has been largely overlooked, he adds, is the absence of a calcium-deficient cohort. “Mean calcium intake in women entering the WHI was twice the national average for women of the same age. From such data as are available, one would say that such intakes are already at, or very close to, the threshold point for calcium. Thus, there is no low-calcium contrast group in the WHI, and hence it is impossible to make a statement about whether calcium would or would not be beneficial.”
Dr Heaney drew an analogy to “giving iron to people who don’t need it, and noting that it did not raise hemoglobin. You might wonder why such an obvious point could be overlooked in the reporting of the data of WHI, and the best explanation I know of is that physicians, and clinical investigators particularly, tend to think of nutrients as if they were drugs. They are not, and the 2 don’t behave the same.”
Coinvestigator JoAnn E. Manson, MD, DrPh, chief, Division of Preven?tive Medicine, Brigham and Women’s Hos?pital, Boston, agreed that the study’s results and the media reports about them were out of sync. She told IMWR that WHI did demonstrate the positive effects of supplementation.
“We found evidence of improved hip bone density in the women randomized to calcium/vitamin D com??pared with placebo. Women who adhered to study medication (took at least 80% of their pills) had a significant 29% reduction in risk of hip ?fracture, and women aged 60 and older had a significant 21% reduction in risk,” she emphasized. “There was also a suggestion that women with lower ?calcium at baseline had reduced risk with supplementation, while women who were already replete had no ?evidence of benefit.”
Dr Heaney postulated that the 29% to 30% reduction in hip fracture reported in compliant patients “may well reflect that fraction of the group that had intakes below average. They are the ones in WHI who might be expected to respond.”
Drs Heaney and Manson agreed that although the average calcium intake among the WHI participants either met or exceeded recommended levels, there was no benefit reported in women who took more than 1000 mg/day. This can be easily explained, according to Dr Heaney: “It exhibits what is called ‘threshold behavior.’ That is, benefits accrue up to some intake, the threshold, above which nothing further happens.” Returning to his earlier analogy, he added, “Giving iron to an iron-deficient woman will raise hemoglobin, but giving more iron after the hemoglobin gets up to the right level will not produce more blood.”
Although the calcium dose may have been sufficient, the vitamin D dose was much too low, according to Dr Heaney. “Typical osteoporotic patients today will be given 1000 to 2000 IU/day, not the 400 IU/day that was built into the WHI design (let alone 240 IU, which is what the intake amounted to, factoring in compliance).”
Walter Willet, MD, Fredrick John Stare Professor of Epidemiology and Nutrition, Harvard School of Public Health, agrees that the low dose of vitamin D used in the trial was one of its “severest limitations.”
“More recent studies, including a meta-analysis that we published last year [JAMA. 2005;293: 2257-2264], have shown that 400 IU/day is unlikely to have much benefit for fracture reduction, but that 700 to 800 IU/day does. It is likely that higher doses will add further reduction in fracture risk,” Dr Willet told IMWR.
Another limitation of the study was compliance, Dr Willet said. “Only 60% of the women were taking their assigned pills during the study, and by the end of the study, 69% were taking calcium on their own, and about one third were taking multivitamins containing vitamin D on their own. It is almost unheard of that in a trial there are more drop-ins than there are people actually taking the study preparations, but that is exactly what happened in the WHI.”
The second part of the study found no evidence that calcium and vitamin D supplementation helped prevent colorectal cancer. Over an average of 7 years, 322 women were diagnosed with invasive colorectal cancer. There was no significant difference between the treatment and placebo groups in number of polyps reported, cancer cases, or tumor characteristics or severity. Compliance did not make a difference.
Dr Willet notes that because of the major limitations of the WHI trial, which contradicts previous evidence that calcium/vitamin D supplementation conferred protection, “it adds little to our understanding of the health effects of calcium and vitamin D.”
Dr Heaney attributed the lack of perceived benefit in part to the “grossly inadequate” dose of vitamin D used in the study. “Recent dose calculations show that it would have taken at least 2000 IU/day to move people from the middle of the bottom quartile to the middle of the top quartile.”
The Bigger Picture
Drs Heaney, Willet, and Manson concur that supplemental vitamin D and calcium continue to be beneficial, particularly for women with inadequate calcium intake or low vitamin D status.
Improvement in bone density usually, although not always, translates into a reduced risk of fractures, especially among older women at risk for osteoporotic fractures, notes Dr Manson. “Thus, increased bone density can be viewed as a form of insurance, which should eventually pay dividends in reducing fracture risk.”
Supplemental calcium and vitamin D has been linked to a lower risk of several other chronic diseases, including cancer (Int J Fertil Womens Med. 2005; 50:244-249), diabetes (see page 15, this issue), hypertension (Am J Clin Nutr. 2004;80:1678S-1688S), and multiple sclerosis (Neurology. 2004;62:60-65), according to Drs Willet and Manson.
Because of other evidence indicating that most Americans have suboptimal vitamin D levels, Dr Willet suggests that most healthy adults should take 1000 IU/ day. “This is a conservative amount, and higher intakes may prove to be? better.”
The situation for calcium is more complicated, he notes, and the generally recommended intake (1200 mg/d after age 50) is probably higher than needed. “I think the United Kingdom definition of adequate intake of 700 mg/day for all persons over 19 years of age is reasonable,” Dr Willet told IMWR.