Renal Function and Risk of Hip and Vertebral Fractures in Older Women: Title and subTitle BreakIs It Always Osteoporosis?

Chronic kidney disease (CKD) is rapidly becoming a significant public health problem. A staggering 19.2 million Americans, or 11% of the adult population, are currently living with CKD.¹ Moreover, there are nearly 400 000 people in the United States with end-stage kidney disease (ESKD), a number that is almost certain to rise owing to the high prevalence of CKD.² Adding to the morbidity of CKD is the development of disturbances of mineral metabolism, which occurs in virtually all patients during the progression of their disease and is associated with bone loss and fractures, cardiovascular disease, immune suppression, and increased mortality.³

In addition to the public health impact, CKD and ESKD also impose a tremendous economic burden. In fact, the cost to treat patients with ESKD in the United States was estimated to be more than $25 billion in 2002.⁴ While dialysis treatment accounts for the majority of these costs, hip and vertebral fractures contribute to the excessive economic burden.⁵ Because patients with CKD and ESKD have reduced bone mineral density (BMD) in association with several metabolic disturbances, in particular hyperparathyroidism and vitamin D deficiency, it stands to reason that they are at increased risk of hip fracture. Although essentially all patients with CKD have decreased BMD, it has not been a useful tool to determine fracture risk in this patient population.

While several studies have shown an increased risk of hip fracture in patients with CKD who are undergoing dialysis, to my knowledge the existence of a similar correlation has not previously been explored in patients with milder stages of CKD. In this issue of the ARCHIVES, Ensrud et al⁶ report the findings of a prospective case-cohort study in which they examined the association of hip fracture in patients with mild to moderate impairments in renal function. Within a cohort of more than 9700 women 65 years or older, they compared the baseline kidney function in women who subsequently had either hip or vertebral fractures with that in randomly selected matched controls. Patients were sorted into 1 of 3 groups based on estimated glomerular filtration rate (eGFR): moderate decrease (<45 mL/min per 1.73 m²), mild decrease (45-59 mL/min per 1.73 m²), and normal (>60 mL/min per 1.73 m²). When the GFR was estimated using the Cockcroft-Gault method, there was a significant increase in the hip fracture hazards ratio in women with a below-normal eGFR. Moreover, this increased hazards ratio was observed even after adjustment for age, weight, and either calcaneal or femoral neck BMD (P = .02). Interestingly, no significant increase was observed in the vertebral fracture hazards ratio (P = .04).

The investigators' choice of equations for determining the eGFR significantly affected the findings. In fact, when Ensrud and colleagues calculated the eGFR using the 4-variable version of the Modification of Diet in Renal Disease index,⁷ the hazard ratios were smaller in magnitude and did not reach statistical significance (P = .09). While it is concerning, as the authors state, that 2 different methods yielded different results, previous studies have shown that limitations associated with the Modification of Diet in Renal Disease index include underestimation of the basal GFR and inaccuracies when the GFR is greater than 60 mL/min per 1.73 m² as well as when individuals are at the extremes in weight.⁸ Thus, an underestimation of the GFR would incorrectly place some patients with a “normal” GFR into either the mild or the moderate decrease eGFR group.

Furthermore, the determination of the independent impact of CKD on fractures was compromised, as Ensrud and colleagues selected a population of women with a high prevalence of postmenopausal osteoporosis. Although it has been demonstrated that patients with CKD have decreased BMD,³ the value of BMD in the evaluation of the bone diseases associated with CKD is not well established.⁹ Findings on the correlation of BMD values to fracture risk in the CKD population are inconsistent.¹⁰ Therefore, the decreased bone mass and increased bone fragility in this particular population are most likely the result of the relative contribution of postmenopausal osteoporosis and the metabolic disturbances associated with CKD.⁹ There are several disorders in CKD, independent of osteoporosis, that may contribute to an increased risk of hip fracture. Specifically, the development of secondary hyperparathyroidism and vitamin D deficiency will lead not only to bone loss but also to defects in bone quality.^{3 ,9 ,11} In addition, the elevated circulating concentrations of fibroblast growth factor 23, which is produced to maintain normophosphatemia, further inhibit renal 1α-hydroxylase activity and may also have a direct detrimental effect on bone quality.^{3 ,12 - 13} Although metabolic disturbances were not independently associated with increased hip fractures in Ensrud and colleagues' analysis, an earlier study, using patients from the same cohort, that did not specifically consider CKD found an association between reduced serum 1,25-dihydroxyvitamin D and increased risk of hip fracture.¹⁴ Clearly, the pathogenesis of bone disease in patients with CKD is extremely heterogeneous, and it is not surprising that any single factor would be predictive of fracture in this patient population. To address the complexity associated with the bone and mineral disorders associated with CKD, an international consensus conference was held in October 2005 under the auspices of Kidney Disease: Improving Global Outcomes (www.kdigo.org) to define and classify “renal osteodystrophy.”¹⁰ This expert panel concluded that because the manifestations of mineral and bone abnormalities were so diverse and included extraskeletal manifestations, a new systemic disorder should be defined and should be called CKD–mineral and bone disorder. The diagnosis of CKD–mineral and bone disorder would include patients who manifest one or a combination of the following symptoms: (1) abnormalities of calcium, phosphorus, parathyroid hormone, or vitamin D metabolism; (2) abnormalities in bone turnover, mineralization, volume, linear growth, or strength; and (3) vascular or other soft tissue calcification.¹⁰

Therefore, the findings reported by Ensrud et al⁶ are potentially very important, as they further support the concept that a diagnosis of osteoporosis based on BMD criteria should not be made in patients with CKD and used as a predictor of fracture outcome. Because of the rapidly increasing prevalence of CKD, it is virtually guaranteed that there will be a corresponding increase in the incidence of CKD-associated hip fractures. Importantly, the mortality associated with hip fracture ranges from 13% to 37%, with only 60% of surviving patients returning to their prefracture level of walking.^{15 - 17} Moreover, the estimated cost to treat a single hip fracture is more than $30 000.⁵ Thus, it is extremely important to understand the pathophysiologic mechanisms that cause CKD-associated fractures and to realize that therapy should be directed toward the underlying disorders. Because patients with CKD have an extremely high prevalence of hyperparathyroidism and vitamin D deficiency, effective treatments that can correct these disorders may reduce the burden of fractures in patients with CKD. Several active vitamin D compounds, such as calcitriol and paricalcitol, as well as the prohormones alfacalcidol and doxercalciferol have been shown to significantly reduce parathyroid hormone levels and moderate the symptoms that are commonly associated with hyperparathyroidism.^{11 ,18 - 20} It has been shown that treatment with the prohormone alfacalcidol improves BMD while reducing parathyroid hormone concentrations in patients with CKD.¹¹ Moreover, it has also been shown that administration of paricalcitol prevents CKD-associated decreases in BMD in the femoral neck and the femoral midshaft and restores bone strength in the femoral neck in experimental animals.²¹ Thus, while mild to moderate kidney impairment puts patients at an increased risk for hip fracture, proper control of hyperparathyroidism and vitamin D deficiency has the potential to reduce the risk of CKD-associated hip fractures.

In summary, it is important to recognize that the risk of fracture in patients with CKD is not just a function of having decreased bone density or of being given the diagnostic label of osteoporosis. It is important to recognize that in the presence of CKD the underlying pathogenesis may be the result of altered mineral metabolism associated with CKD-MBD. Therefore, before antiosteoporosis therapy is initiated, these patients should be further evaluated and treated for any existing underlying disorders.

Correspondence: Dr Sprague, Evanston Northwestern Healthcare, Northwestern University Feinberg School of Medicine, Division of Nephrology and Hypertension, 2650 Ridge Ave, Evanston, IL 60201 (ssprague@northwestern.edu).

Financial Disclosure: None reported.