Letter to the Editor: The Endocrine Society Clinical Practice Guidelines on Paget's Disease: Many Recommendations Are Not Evidence Based

Dear Sir, We write with regard to the Endocrine Society Clinical Guideline on Paget's disease (1). Preparing guidelines that balance the available evidence with the need to give clinicians practical advice is a difficult task, especially in conditions like Paget's disease of bone (PDB) where high-quality evidence on patient-centered outcomes is lacking. Notwithstanding the challenge the authors faced, we have concerns that many recommendations seem to have been based on personal opinion rather than evidence and that some relevant evidence has been overlooked. In the background, the authors state that bone pain is only present in a minority of patients with PDB. Although it is true that many patients with PDB never come to medical attention (2), the guideline focused on the management of those that do present clinically. In this population, bone pain occurs in about 50% of cases (3). In view of this, we were surprised that treatment of bone pain did not figure more prominently in the guideline because the evidence base in support of this is strong (4). Recommendation 2.1 (moderate quality evidence) advises that most patients with active PDB who are at risk of complications should be treated with a bisphosphonate. This recommendation is impossible to implement because there is no validated method of identifying PDB patients who are at risk of complications. Recommendation 2.5 (low quality evidence) suggests that bisphosphonate treatment should be given with the aim of reducing the chosen biochemical marker to below the midpoint of the reference range. The implication from both of these recommendations is that bisphosphonate treatment in general and normalizing bone turnover in particular are beneficial in PDB in helping to prevent complications. However, the supporting evidence is that “it seems reasonable to suppose despite the lack of objective evidence that long-term complications might be reduced by normalization of bone turnover.” With all due respect to the authors, this is an opinion and not evidence. A randomized trial has been performed to investigate whether it is beneficial to try and normalize bone turnover in PDB, but no difference was found in the rate of complications between groups of patients treated with the aim of normalizing alkaline phosphatase concentrations, as opposed to those who were treated symptomatically (5). Recommendation 2.6 (moderate quality evidence) similarly suggests that biochemical markers of bone turnover should be used as a more objective indication of relapse than symptoms. Implicit in this recommendation is that “biochemical relapse” predicts clinical relapse. We are aware of no evidence to suggest that this is the case. Recommendation 3.1 (low quality evidence) suggests that a potent bisphosphonate should be given to prevent worsening of a hearing deficit, with supporting evidence from an observational study with calcitonin (6). The effects of bisphosphonates on hearing loss in PDB have been studied, and no benefit has been observed (5, 7). Recommendation 3.2b (low quality evidence) suggests that bisphosphonates should be given to prevent cartilage degeneration in patients with osteoarthritis adjacent to an affected bone. The supporting evidence is that “it is conceivable that drug treatment of Paget's disease may slow the arthritis process.” This recommendation is based on the authors' opinion, rather than evidence. We have great respect for the authors of the guideline, many of whom have extensive personal experience in the treatment of PDB. It is crucially important, however, that clinical guidelines accurately portray the level of supporting evidence, and if no evidence exists, this should be highlighted as a gap in knowledge. These guidelines have not done that adequately.