Eli Lilly and Company, Indianapolis, Indiana 46285
Address all correspondence and requests for reprints to: Robert Marcus, M.D., Eli Lilly and Company, 1400 Raymond Street, Building 172, Indianapolis, Indiana 46285.
Dialogue regarding diagnostic testing in patients who have or are at risk for osteoporosis has focused largely on the assessment of bone mass and turnover. Questions raised have included dual energy x-ray absorptiometry vs. other methods, central vs. peripheral testing, the value of quantitative ultrasound, and the use of bone turnover markers for individual patients. However, relatively little attention has been paid to the medical evaluation of patients with osteoporosis. Better understanding of the influences and disorders that affect skeletal health has led to the evolution of a term, "secondary osteoporosis," that includes a laundry list of conditions linked to deterioration of skeletal microarchitecture. Current texts offer diagnostic approaches aimed at discovering secondary forms of osteoporosis, some of which are rare, but these strategies most often reflect accumulated clinical wisdom without the benefit of formal validation. Consequently, a consensus does not yet exist as to which tests can be done in a cost-effective manner that would justify their routine use.
In this issue of JCEM, Tannenbaum et al. (1) report on their experience with laboratory testing of otherwise healthy women who were seen at a large urban osteoporosis referral center. Among the 173 women who met inclusion criteria for this study, almost half had at least one abnormal test result, leading to the identification of a previously undiagnosed disorder of bone and mineral metabolism in 55 women, or one third of the study sample. The most commonly encountered diagnoses were related to the calcium-parathyroid-vitamin D axis, representing both primary and secondary hyperparathyroidism, various forms of hypercalciuria, and intestinal malabsorption. Of particular interest, clinical risk factors for the presence of one of these disorders were notably absent. Only a personal history of localized breast cancer and of smoking independently predicted the presence of a disorder of bone and mineral metabolism. Thus, the only secure way to exclude such disorders is by routine laboratory testing.
This thoughtful analysis makes a strong case for routine examination of the calcium-parathyroid-vitamin D axis in women with osteoporosis. The reasons for doing so are multiple. Currently approved antiresorptive drugs offer improvement in bone mineral density and reduce the risk of fracture, but their success is predicated on the presence of an osteoporotic skeleton unencumbered by defective mineralization (osteomalacia), hyperparathyroidism, or intestinal malabsorption. Occult malabsorption syndromes are of particular concern in this regard, not only because the undermineralized skeleton may respond poorly to therapy, but also because of the potential impact that intestinal mucosal dysfunction might have on drug absorption.
The work of Tannenbaum et al. (1) raises several issues that merit discussion. First, their experience derives from an almost exclusively Caucasian tertiary clinic practice in an academic medical center. The general applicability of these results remains unknown. Second, of more than 600 women screened, 355 failed to qualify for the study because of a history of exposure to one or more medications or diseases that themselves are known to disrupt bone and mineral metabolism. Thus, by adding the 55 women with a newly diagnosed disorder to the 355 whose underlying conditions were previously identified, one sees that more than 60% of patients reporting to an osteoporosis clinic have elements of secondary osteoporosis contributing to their diagnosis.
A central feature of the authors diagnostic strategy is the evaluation of 24-h urinary calcium excretion. Tannenbaum et al. (1) report that such testing assisted in the detection of several disorders, including hyperparathyroidism, hypercalciuria, and malabsorption syndromes. However, collection of satisfactory urine specimens can be problematic. For example, serial urinary calcium determination was at one time recommended as a routine procedure in the management of patients with primary hyperparathyroidism. This recommendation has been abandoned in part because complete collections were difficult to obtain (2). As noted by Tannenbaum et al. (1), 24-h urinary calcium excretion varies substantially with degree of estrogen repletion and with dietary intake of protein, sodium, and other nutrients. Whereas accounting for these variables may be feasible in a large endocrine or osteoporosis clinic, its practicality in a demanding primary care setting is less clear.
The use of urinary calcium to detect celiac disease is even less certain. In the cohort reported by Tannenbaum et al. (1), three patients with hypocalciuria subsequently were found to have nontropical sprue. Evidence in the literature to support routine use of urinary calcium excretion in the diagnosis of celiac disease is limited. Short of intestinal biopsy, determination of antigliadin and antiendomysial antibodies is the laboratory standard (3). The cost of a 24-h urinary calcium determination roughly equals that of antibody evaluation. However, variable sensitivity of the antibody assays introduces sufficient false positive findings that additional costs would be incurred (3). More studies like that of Tannenbaum et al. (1) would help clarify the effectiveness of urinary calcium as a screening tool for celiac disease.
Tannenbaum et al. (1) developed cost estimates for several laboratory strategies aimed at identifying the greatest number of patients with these common secondary factors. They concluded that obtaining measurements of 24-h urinary calcium excretion and serum concentrations of calcium, PTH, and TSH identified 85% of affected women at a nominal cost of $75 per patient.
Although vitamin D insufficiency was found to be a prominent contributor to their results, Tannenbaum et al. (1) considered the incremental value of measuring the 25- hydroxyvitamin D (25OHD) concentration to be modest and have not included it in their recommendations for screening. Because they used a conservative approach to assess vitamin D status, it is very likely that the true prevalence of marginal vitamin D nutritional state was considerably higher than stated. Only individuals with serum 25OHD concentrations below 30 nmol/liter (12.5 ng/ml) were considered to be low. Although bone mineralization deficits (osteomalacia) are associated with 25OHD values below 10 ng/ml, it is now clear that values of 3075 nmol/liter (12.530 ng/ml) are associated with compensatory hypersecretion of PTH, increased bone turnover, and aggravation of bone loss (4). With increased recognition of widespread vitamin D inadequacy in the older population, and growing acceptance that full vitamin D repletion requires 25OHD concentrations of 75 nmol/liter (30 ng/ml) or greater, a case can still be made for assessing 25OHD as a routine part of the evaluation of patients with osteoporosis.
The state of ones skeletal health represents the cumulative effect of myriad influences that impinge on bone throughout life (the phrase "you are what you eat" comes to mind). Such influences include those affecting bone acquisition, such as genetic endowment, childhood nutrition, and physical activity, and adult exposures, such as disease, immobility, and medications, including some that were exclusionary for Tannenbaum et al. (1). Some forms of osteoporosis warrant full-fledged status as independent clinical entities. These include conditions in which the mechanisms of bone loss, histological appearance, and skeletal distribution, are unique. Examples include systemic mastocytosis and probably glucocorticoid-associated osteoporosis. However, for most other secondary osteoporoses, it is more appropriate to consider the putative inciting factor just another "hit" among many that have been experienced over a lifetime.
By the time a patient pursues evaluation for osteoporosis, she or he likely has undergone clinical chemistry testing as part of routine care. Results of these tests suggesting the presence of an underlying medical disorder (for example, abnormal liver function, anemia, or hypercalcemia) can sharpen the focus of further inquiry and are, therefore, cost-effective. Tannenbaum et al. (1) propose an intriguing additional strategy to assess a variety of underlying medical conditions that would not be diagnosed by routine clinical chemistry alone. We congratulate them on initiating a dialogue in this important area and trust that future studies of this sort will lead to the emergence of a consensus for appropriate diagnostic testing of patients with osteoporosis.
Acknowledgments
Footnotes
Abbreviation: 25OHD, 25-Hydroxyvitamin D.
Received August 16, 2002.
Accepted August 18, 2002.
References