Renal osteodystrophy, the spectrum of bone disorders that develops in patients with chronic renal failure, continues to be a major long-term complication which is associated with high rates of morbidity [1,2]. Although bone histomorphometry is the most reliable diagnostic method, in recent years several new biochemical markers of bone turnover have been proposed for the non-invasive evaluation of bone remodelling in uraemic patients. The ideal biochemical marker of bone turnover should be unique to bone, reflect total skeletal activity and be well correlated with histomorphometric findings and results of radiocalcium kinetics. We are still looking for a specific and sensitive serum biochemical test for monitoring bone turnover in uraemia. Several enzymes and matrix proteins synthesized by the osteoblast, and protein fragments released after bone matrix breakdown, have been proposed as circulating biochemical markers of bone formation and bone resorption (Table 1). The interpretation of serum concentrations of these markers is hindered, in chronic renal failure, by different factors, including circadian rhythms, diet, age, gender, menopause, liver function and the degree of reduction of renal function (Table 2). Biochemical serum markers of bone turnovers in renal osteodystrophy Biochemical serum markers of bone turnovers in renal osteodystrophy Biological markers of bone turnover Biological markers of bone turnover The development of highly precise radioimmunometric assays for intact parathyroid hormone (iPTH) in the past decade has shown that serum iPTH level concentration is a useful predictor of the type of bone histology. Such assays are useful for non-invasive distinction of high-turnover bone disease (HTBD) from normal bone or low-turnover bone disease (LTBD), at least when entire groups of patients are considered [3]. However, in an individual patient, it is often not possible to use serum iPTH levels alone to clearly distinguish adynamic or normal bone from hyperparathyroid bone disease, as has been demonstrated by Qi et al. [4]. Those authors found that serum iPTH levels between 65 and 450 pg/ml were not predictive of the degree of bone turnover in dialysis patients; invasive diagnosis was proposed for these patients. The Toronto Renal Osteodystrophy Study evaluated 256 chronic dialysis patients, measuring their plasma 1–84 PTH levels and assessing bone histology. Although patients with HTBD generally had high 1–84 PTH values, nine of these patients had levels of <200pg/ml (and two <100pg/ml). In four of these nine patients, bone histology showed significant trabecular surface aluminium staining. Such patients are an important source of diagnostic uncertainty, which can only be resolved by bone biopsy [5]. iPTH levels vary according to the type of dialysis (haemodialysis versus CAPD) [6], the degree of aluminium overload [5-7] and probably other factors, so different iPTH cut-off levels have been proposed for different patient populations, to optimize the sensitivity and specificity of serum iPTH concentration as a diagnostic tool (Figure 1). Predictive value of serum iPTH levels for bone turnover in dialysis patients. Recent data by Gerakis et al. [8] show that in patients with end stage renal disease, plasma iPTH concentrations of < 65 pg/ml have a poor positive predictive value (45%) in the diagnosis of adynamic bone disease, but have an excellent negative predictive value, even in patients with significant bone aluminium deposition. The same authors showed that a serum iPTH concentration 3.5 times above the upper limit of normal had a positive predictive value of 97% in the diagnosis of HTBD; this positive predictive value dropped to 78% in the presence of positive aluminium staining. In a study of patients from Portugal and Spain with adynamic bone disease, solochrome azurine staining was used to assess aluminium deposition in bone; twothirds of the patients had > 25% stainable bone surface, showing that aluminium still plays a role in the pathogenesis of adynamic bone disease [9]. The interpretation and significance of iPTH levels in the individual patient is further complicated by other factors, such as the proposed skeletal resistance to PTH in chronic renal failure, as postulated by Massry et al. in 1973 [10] and recently confirmed by our group at the level of the epiphyseal cartilage growth plate in uraemic rats [11]. Recently, Lepage et al. showed that in most of twosite iPTH assays there is cross-reaction with the non-(1-84) molecular form of PTH, accounting for 40–60% of the measured immunoreactivity. These results could explain, at least in part, the higher iPTH concentrations measured in plasma from uraemic patients [12]. It is thus clear that additional tests or new markers of bone remodelling are needed in order to obtain correct and dynamic non-invasive diagnosis of bone turnover. Such additional biochemical markers will have to be evaluated to determine their diagnostic value. The majority of recent studies using serum markers other than iPTH are listed in Table 1 and discussed below. Alkaline phosphatase (AP) is a glycosylated protein produced by at least five different organs, namely liver, bone, kidney, intestine and placenta. In uraemia the relative contribution of each of these isoenzymes to serum total AP activity may be modified by changes in their relative half-lives [13]. In bone, AP is produced by osteoblasts and osteoblast precursors and participates in the mineralization process. The development of monoclonal antibodies specific for bone alkaline phosphatase (bAP) provided the basis for a more specific index of bone formation [14]. In a recent study involving 42 haemodialysis patients, we showed that plasma bAP levels (measured with a new direct radioimmunometric assay) were better correlated with histomorphometric parameters of bone formation and bone resorption than plasma iPTH or total AP levels [15]. Values of bAP >20 ng/ml had a sensitivity of 100% and a specificity of 100% for the diagnosis of HTBD and there was an excellent correlation between plasma bAP levels and bone formation rate. When these limits of bAP values were associated with iPTH serum levels > 200 pg/ml, the positive predictive value for the diagnosis of HTBD increased from 84% to 94%. These results have been confirmed by other groups in HTBD patients [16]. The usefulness of bAP as a marker of adynamic bone disease has recently been proposed by Couttenye et al. [17], using an agarose gel electrophoresis method for detecting low plasma concentrations of bAP. These authors demonstrated that a low (<27 U/l) level of bAP and a low (<150 pg/ml) level of iPTH are good markers of adynamic bone disease with sensitivities of 78.1% and 80.6% and specificities of 86.4% and 76.2%, respectively. The same group, using these criteria, showed a high prevalence of adynamic bone disease (43%) in 212 European CAPD patients [18]. Bone alkaline phosphatase also showed the best diagnostic reliability as a marker of bone formation, in relation to length of time on dialysis [19]. It still needs to be demonstrated in a larger patient population that bAP is sensitive enough to distinguish between low and normal bone turnover. Recent results by Fletcher et al. [20] suggested that the non-invasive markers of bone turnover used by them (iPTH, total AP and bAP) could not be used to distinguish between low-turnover bone disease, normal bone and mild osteitis fibrosa. The use of bAP has to be balanced against the current lower availability and higher cost of the assay compared with AP. Osteocalcin (BGP) is one of the most abundant noncollagenous proteins in bone and is also present in dentin and calcified cartilage. It is produced by osteoblasts and has long been considered to be a marker of bone formation. Serum levels of intact osteocalcin seem to reflect the excess of produced protein that has not been integrated into bone matrix, or resulting from bone resorption. The new radioimmunometric assays, which use specific monoclonal antibodies to measure only intact osteocalcin [21], may be more accurate, since they should not measure osteocalcin fragments. However, plasma osteocalcin suffers from poor stability and is removed via the kidneys. After the menopause its serum concentrations increase less than those of bAP, so it is less sensitive in the diagnosis of bone turnover in this type of bone disease [22]. Procollagen type I carboxy-terminal propeptides (PICPs) have been used as a marker of bone formation since they are by-products of collagen synthesis, and collagen type I is the most abundant protein of bone, accounting for > 90% of the proteins of bone matrix. In pre-dialysis chronic renal failure, PICPs showed a high correlation with dynamic, but not with static, histomorphometric parameters or with other humoral markers of bone turnover [23]. We found increased levels of PICPs in our haemodialysis patients, but we did not observe a significant correlation with any of the histomorphometric parameters examined [24], in accordance with the results of Mazaferro et al. [25]. In a population of 18 haemodialysis patients, Hamdy et al. [26] found inappropriately increased PICP levels in patients with aluminium overload (38% of the patients studied) whereas serum total AP and osteocalcin levels were decreased, also indicating a relatively low specificity of PICPs as diagnostic markers in such patients. This is the fraction of acid phosphatase that shows resistance, as judged by its mobility on acrylamide gel, to inhibition by tartrate [27]. The recent development of immunoassays for this enzyme should offer increased specificity, but there are still few clinical data [28, 29]. Type I collagen cross-linked telopeptide (ICTP) is a collagen derivative that is released during bone resorption. Assays for ICTP using a polyclonal antibody have been developed [30] but with disappointing clinical results [31]. The rate of ICTP elimination depends on glomerular filtration rate and ICTP serum levels respond poorly to hormone replacement therapy after the menopause [32]. In our dialysis patients, we failed to observe a significant correlation between serum levels of ICTP and any of the numerous static or dynamic parameters provided by histomorphometry, suggesting that ICTP is not a sensitive marker of bone metabolism in uraemia [26]. In contrast, Mazzaferro et al. found that serum ICTP levels were correlated with total AP, bAP, and iPTH serum concentrations and with several histomorphometric indices of bone turnover, and they proposed ICTP as a useful humoral marker of bone turnover in renal osteodystrophy [25]. Pyridinoline cross-links of collagen exist in two chemical forms, namely hydroxylysyl pyridinoline (or pyridinoline (Pyr)) and lysyl pyridinoline (or deoxypyridinoline (Dpy)). These molecules are markers of type I and II collagen breakdown and their excretion in urine has already been validated as an excellent marker of bone resorption in several metabolic bone diseases. Recently, we have demonstrated for the first time that serum Pyr concentrations can be reliably measured with an accurate competitive enzyme immunoassay [33] in dialysis patients, and that these patients had markedly higher serum Pyr levels than normal individuals. In our 37 haemodialysis patients, the highest values of serum Pyr were found in patients with the highest rate of bone resorption. Moreover, we observed a good correlation between serum Pyr levels and the number of osteoclasts/mm2 as well as the proportion of bone covered by osteoclasts [24]. β2-Microglobulin (β2M), a 11 810 Da polypeptide, is the main constituent of amyloid deposits found in osteoarticular structures of haemodialysis patients. Since the catabolism of β2M takes place mainly in the kidney, a progressive increase in β2M serum levels is observed with declining renal function, and very high levels are observed in anuric patients with end-stage renal failure [34], ranging from 30 to 60 mg/l (much higher than the normal value of 0.8–3.0 mg/l). However, elevated β2M levels alone do not define the diagnosis of dialysis-related amyloidosis, since they may be seen in patients without bone disease. Recently, we observed that patients with histological HTBD had higher serum β2M levels than patients with N/LTBD [35] and that serum β2M levels were correlated with two serum markers of bone formation rate, namely osteocalcin and bAP, and also with serum free pyridinoline (a specific serum marker of bone resorption), but not with serum iPTH. Lastly, serum β2M levels were negatively correlated with osteoid volume in patients with HTBD. The association of high serum β2M with higher bone cell number and with serum markers of bone turnover suggests that β2M could be either a direct or an indirect activator of bone cells or at least another marker of bone cell activity. At present, the non-invasive evaluation of bone turnover should include a combination of different markers, so that the balance between bone formation and bone resorption can be evaluated adequately. Besides plasma iPTH, the evaluation by immunoassay of human bAP for bone formation, and of pyridinoline for bone resorption, appear to be the most sensitive and specific serum markers of bone turnover. These markers should be combined with serum iPTH, β2M and aluminium levels and be monitored according to the schedule shown in Table 3. Suggested schedule for biochemical markers Suggested schedule for biochemical markers In conclusion, many new biochemical markers of bone metabolism have been developed in recent years as diagnostic tools for the non-invasive diagnosis of bone disease. In the area of renal osteodystrophy, we are still waiting for the ideal biochemical marker of bone turnover, since each of the markers discussed above has limited diagnostic usefulness. It must be pointed out here that several authors are not convinced that these new biochemical markers allow more accurate diagnosis of each subtype of renal osteodystrophy than the combined use of iPTH and aluminium alone. The management of each patient should be individualized; bone biopsy is indicated whenever diagnosis is equivocal or the patient's symptoms are not consistent with laboratory findings (see below, ‘Indications for invasive diagnosis’). Bone histology remains the gold standard for the diagnosis of renal osteodystrophy. Precise evaluation of the underlying type of bone disease in these patients frequently requires bone histomorphometry (including static and kinetic variables after double tetracycline labelling), which is an invasive and costly method. There are four major categories of bone disease according to histological evaluation: osteitis fibrosa and mixed osteodystrophy, which are both HTBDs, and osteomalacia and adynamic bone disease, which are LTBDs. Osteitis fibrosa, which is predominantly the histological expression of secondary hyperparathyroidism, is characterized by a high bone formation rate, namely a marked high extension of the trabecular surfaces presenting single or double tetracycline labels, and an increased number of osteoblasts, osteoclasts and eroded surfaces. A marked increase in bone marrow fibrosis and osteoid surface is usually also present. Mixed uraemic osteodystrophy is characterized by signs of secondary hyperparathyroidism with, in addition, increased volume and thickness of osteoid volume. Osteomalacia is characterized by a low bone formation rate associated with an increase in surface, volume and thickness of osteoid. Adynamic bone disease is characterized by defective bone formation, together with a marked decrease in active remodelling sites and a low bone formation rate. The prevalence of adynamic bone disease has probably increased during the last decade in dialysis patients, to reach a proportion varying between 15% and 60% of patients [36,37]. The width of this range can probably be attributed to a number of factors, including varying patient selection criteria, differences in the prescription of aluminium- vs calcium-based phosphate binders, inappropriate treatment of tapwater and variable usage of vitamin D metabolites. The prevalence of osteomalacia and of mixed uraemic osteodystrophy is decreasing, and both diagnoses are made in < 10% of bone biopsies. The most relevant approach to histological diagnosis is to distinguish between HTBD and LTBD and then define the mineralization process and quantify the eventual aluminium deposits. Although the prevalence of aluminium overload has diminished significantly in industrialized countries, it remains the most frequent cause of a primary defect in bone mineralization and of adynamic bone disease in Latin-American countries, as has been recently shown in a review evaluating 1209 bone biopsies from symptomatic patients [9]. In this study, approximately 20% of patients had adynamic bone disease (similar to the proportion found in previous extensive surveys [38,39]) but two-thirds of patients from Portugal and Spain with adynamic bone disease had a solochromestainable bone surface of > 25%. These results justify the frequent need for bone biopsy in such countries to determine the exact type of skeletal lesion whenever there is a suspicion of significant aluminium exposure [40]. It is important to diagnose the type of bone lesion correctly so that appropriate treatment can be given since different lesions need different, and frequently conflicting, therapy. The suspicion of aluminium overload may arise from the patient's dialysis history (water treatment and type of phosphate binders used) and from plasma aluminium levels. The latter have to be interpreted simultaneously with the iron status (measurement of plasma ferritin and transferrin saturation [41]), because patients with iron deficiency may exhibit high serum aluminium levels in the presence of only a modest increase in the burden of aluminium, as evaluated by a desferrioxamine test [42]. On the other hand, patients with high ferritin levels (>1000 mg/l ) may have a positive desferrioxamine test and significant bone accumulation of aluminium and yet exhibit serum aluminium levels of <30 mg/l [43,44]. Ideally, all patients in a dialysis unit should have plasma aluminium levels of <20–30 mg/l. Plasma aluminium has been measured after a standardized infusion of desferrioxamine to assess tissue stores of aluminium. D'Haese et al. evaluated the diagnostic value of a positive low desferrioxamine test (> 50 mg/l increase in plasma aluminium after administration of 5 mg/kg of desferrioxamine), used in conjunction with serum iPTH levels, in the diagnosis of aluminium-related bone disease [45]. Patients with an increase in the serum aluminium level of > 50 mg/l and having a plasma iPTH of > 650 pg/ml have aluminium overload, but they are not prone to develop aluminium bone disease. Patients with the same increment in serum aluminium, but a plasma iPTH level of <650 pg/ml are prone to show an effect of aluminium in inhibiting bone mineralization, while the greatest risk of aluminium toxicity exists in patients with an iPTH level of <150 pg/ml. In the presence of significant aluminium exposure, bone biopsy seems indicated in the following cases: (i) before parathyroidectomy (since aluminium-induced LTBD may be precipitated by correction of the hyperparathyroid state [46,47]) and (ii) before starting longterm desferrioxamine treatment, as this has risks (i.e. deafness, fatal mucormycosis [31]). Besides these indications for an invasive diagnosis, bone biopsy may have to be performed to establish the diagnosis of adynamic bone disease in symptomatic patients. In our opinion, bone biopsy can also be very useful in directing therapy in some symptomatic patients with intermediate plasma PTH values (between 100 and 450 pg/ml), when the diagnosis is equivocal or when the patient is in transition from a high-turnover state to an adynamic state (or vice versa). The indications for bone biopsy are summarized in Table 4. When is bone biopsy needed? When is bone biopsy needed? Parts of this text were published in Nephrol Dial Transplant 1998; 13 [Suppl 3]: 29-32.
No takes yet. Share an insight, caveat, or question.
A 2000 study studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: