Bone Biopsy in Chronic Kidney Disease: Still Neglected and in Need of Revitalization

Sandro Mazzaferro; Marzia Pasquali


Nephrol Dial Transplant. 2021;36(2):202-204. 

Renal osteodystrophy (ROD) is now included in the larger chronic kidney disease–mineral bone disorder (CKD-MBD) with evidence of a specific contribution to increased mortality in renal diseases. Bone biopsy is universally recognized as the gold standard diagnostic test for ROD, but due to limited diffusion, Kidney Disease: Improving Global Outcome (KDIGO) guidelines do not recommend it widely. The now better-appreciated endocrine role of bone and the intriguing relationship between osteoporosis and ROD in chronic renal failure warrants a definite appreciation of the pathomechanisms involved in these diseases to decide the best therapy. In particular, antiresorptive therapies now available for osteoporosis should be better evaluated in renal patients with bone biopsies. For this reason, the paper by Novel-Catin et al.[1] suggesting the use of a small needle to perform bone biopsies deserves attention, because it could enable the increased use of bone biopsies in CKD.

ROD, the complex and polychrome bone lesions that develop in patients with CKD, is one of the three elements that characterize a disorder, introduced in 2006, and named CKD-MBD.[2] Together with laboratory derangements of divalent ions and vascular and ectopic calcifications, ROD contributes to a complex metabolic disorder with significant systemic effects that are responsible for increased mortality. The growing knowledge of bone cells physiology, which led to the appreciation of bone as an endocrine gland, warrants an improvement of our understanding of the pathomechanisms leading to the development of ROD.

KDIGO guidelines on CKD-MBD, updated in 2017,[3] acknowledge the increased fracture rate of CKD G3a–G5D patients as compared with the general population and revitalize (as compared with previous edition) the diagnostic role of bone mineral density (BMD) to measure this risk. By comparison, the role of bone biopsy seems reduced. In fact, despite still being considered the gold standard test to diagnose ROD, it is recommended only when 'knowledge of the type of ROD will impact treatment decisions' (level of evidence: 'not graded'). Thus, on practical grounds, a bone biopsy procedure can be considered if parathyroid hormone trends are inconsistent, fractures or hypercalcaemia are unexplained, and/or the response to therapy is unclear or associated with progressive BMD decrease. Importantly, in these guidelines, bone biopsy is no longer recommended in CKD patients with BMD values and/or clinical data indicating osteoporosis, before initiating antiresorptive therapies (like bisphosphonates or denosumab). The rationale for this recommendation is that, 'if bone biopsy is not available', antiresorptive therapies could be withheld to patients with high risk of fracture. While this could be reasonable in very early CKD G1–2 stages, in more advanced phases and without a balanced evaluation by an expert nephrologist, a risk exists when prescribing these drugs even in patients with osteomalacia or any other type of ROD, for which conditions they are not still validated. Indeed, the relationship between osteoporosis (defined as mechanically incompetent or fragile bone) and ROD (the complex metabolic derangement of bone in uraemia) is intriguing and still unsettled. If we recognize that any stage of CKD associates with an increased fracture rate, ROD could be a type of osteoporosis.[4] However, while expert physicians of osteoporosis hardly ever question the diagnostic value of BMD, nephrologists recognize the limited sensitivity and specificity of either BMD assessment or of circulating biochemical markers in the diagnostic process of ROD. Therefore, while therapeutic choices might be straightforward for the former specialist, concerns could arise for the latter.

As alluded to previously, recent years have witnessed a total revolution in the appreciation of the biologic role played by bone in our body. Nowadays bone is regarded as a sophisticated endocrine organ (probably the largest in our body), synthesizing a number of possibly still unrecognized hormones, with multiple and underappreciated systemic clinical implications.[5] Accordingly, ROD should be viewed as a complex endocrine disorder that warrants more precise appreciation. Interestingly, the early phases of CKD with the associated adaptive or maladaptive responses of kidneys, parathyroids and bone,[6] and characterized by the disarray of the Fibroblast Growth Factor 23 (FGF23)/Klotho/vitamin D axis,[7] have been suggested to represent a human model of accelerated aging.[8] As such, therapies aiming at restoring the mineral and bone disorders of CKD patients could also improve or worsen the metabolic processes of aging. Thus, the clinical impact of ROD during the whole story of CKD (including the conservative, dialytic and transplant phases) warrants the deepest possible appreciation for the most appropriate therapeutic choice. Therefore, if bone histology is the gold standard for diagnosis, nobody would question that we need to increase the evaluation of bone histology in CKD.

Hypothetically, most nephrologists would agree to perform routine bone biopsies in all CKD patients, were it not for the fact that it is an invasive procedure, there is lack of experience with the procedure of sampling and few labs have specific expertise that guarantees proper evaluation. Indeed, there are practical hurdles to obtaining a bone biopsy. Besides qualitative histologic examination, also quantitative histomorphometric measurement is recommended, which requires an intact bone sample, with adequate amounts of cortical and cancellous bone, unaffected by artefacts (e.g. breakage, fractures, compressions, cellular damage). For this reason, dedicated instruments and sufficiently skilled and experienced operators are necessary. As for the site of sampling, given its accessibility and safety, the iliac crest is invariably preferred, with either a vertical or horizontal approach (Figure 1). Instruments can be manual or electric, each with specific advantages and limitations.[9] In any case, the bone sample diameter is crucial, since the quality of 3 mm diameter samples is significantly lower than those of 5 mm.[9] In general, there is agreement that trephines with an internal diameter of 6–8 mm guarantee better results, and that artefacts increase below 4 mm. On practical grounds, this means that more than a thin haematologic needle (3–4 mm), we need a trepanning surgical instrument (7–8 mm), which makes a significant difference in terms of perceived invasiveness by the patient and possibly by the doctor. Invasiveness is the main reason why bone biopsies are rarely performed in renal patients, despite the clinical relevance now acknowledged for bone dysfunction in CKD and the very low rate of complications. It goes without saying that the fewer bone biopsies are performed, the lower will be the number of skilled operators and of experienced laboratories available.

Figure 1.

With any instrument employed (manual, electric, thin or large), either vertical or horizontal approaches are commonly used to perform a bone biopsy. When using haematologic (thin) needles of 3–4 mm internal diameter, a vertical approach seems better, since a greater number of fields is predictably obtained.

A recent paper published by Novel-Catin et al.[1] compares the accuracy of ROD diagnosis obtained on standard 7.5 mm diameter samples and on halved histological bone sections. After standard evaluation, the authors examined the histologic sections after dividing them lengthwise, graphically, with a permanent felt pen, into two hemi-sections resulting of ~3 mm diameter. The average diameter of halved sections was 2.3 ± 0.3 mm, with a mean number of five analysable fields (at magnification ×10), which was the same as the number of fields analysable from renal biopsies obtained with a thin 4 mm diameter needle and a transiliac (horizontal) approach. By comparison, with a vertical approach the same 4 mm diameter needle provided up to eight fields. Thus, the surface available for histologic quantification was similar between the mimicked halved sections and the real smaller samples. Measurement of static and dynamic histomorphometric parameters yielded almost identical values (r-values between 0.89 and 0.99) with a diagnostic concordance among different ROD types of 92%. Notably, the concordance analysis of ROD types was made on the basis of established histomorphometric thresholds, without any consideration of other biochemical and clinical parameters (unavailable in this retrospective study) that could have modified the diagnosis of borderline cases. In any case, a correct classification of bone disease based on the recommended parameters of turnover, mineralization and volume was possible in the vast majority of cases. As a whole, this article suggests that bone biopsy samples of sufficient quality can be obtained with small haematologic needles, which are definitely less invasive and painful for the patient and certainly more practical for the physician. Also, since the number of fields available was higher with the vertical approach, this type of procedure should be favoured. Indeed, this last finding is in agreement with Malluche and Monier-Faugere's experience[9] favouring the vertical approach with 4 mm diameter needles. The study has limitations. First, the retrospective approach resulted in a distribution of ROD types (prevailing high bone turnover cases) quite different from what is expected in more recent population of patients (prevailing low bone turnover). Thus, the diagnostic concordance should be confirmed in contemporary patients. Second, bone samples were originally obtained with larger trephines, which are expected to produce lower artefacts compared with thinner instruments. Therefore, the number of poor-quality bone histology samples obtained with haematologic needles needs to be better defined and is expected to depend significantly from the expertise of the operator. Despite these considerations, we feel that bone biopsy procedures in CKD should be revitalized, and that the simplest way to achieve this is to render it less invasive and more accessible to patients and operators. Haematologic needles could be a solution that warrants verification. An excellent initiative, led by Pieter Evenepoel, as part of the ERA-EDTA CKD-MBD Working Group activities, is to create a European registry of bone biopsies,[10] which, by encouraging the diffusion of the procedure and by collecting large databases, is expected to allow more sophisticated association analyses and more precise evaluation of the effects of available therapies.

In conclusion, viewing the bone as an endocrine gland infers that renal patients, besides the historical osteitis fibrosa, suffer different types of 'osteitis'. Also, if we consider that FGF23, now the most renowned bone hormone, is only one of the numerous non-collagenous bone proteins whose biologic function is under evaluation, it is easy to envisage that ROD by inducing deep changes in bone turnover and metabolism, could modify the amount and quality of these proteins that, by circulating in the blood, could exert potential endocrine effects. For this reason, the first step seems to increase the number of bone biopsies performed in order to deepen our knowledge of ROD.