Glucosamine is a natural substance that has been widely used for several years as a food supplement to treat people with osteoarthritis.1 One formulation of glucosamine hydrochloride (Alateris – Ransom) has recently become the first (and is currently the only) glucosamine product to be licensed as a medicine in the UK for symptomatic relief of mild to moderate osteoarthritis of the knee. In 2002, we concluded that unlicensed "oral glucosamine sulphate 1,500 mg [daily] probably provides modest symptom relief in patients with osteoarthritis of the knee and its efficacy appears similar to that of NSAIDs".1 Here we update our advice in the light of new evidence and assess the place of the licensed glucosamine hydrochloride tablets.

Up to 1 in 25 people in Europe and the USA have stable angina, with symptoms that may limit function and quality of life.1,2 Beta-blockers are usually used in initial symptomatic treatment, but may cause unwanted effects.3^–6 They are also contraindicated in some patients (e.g. those with uncontrolled heart failure, severe peripheral vascular disease) and should be avoided in patients with asthma or a history of reversible obstructive airways disease or bronchospasm.6 Ivabradine (Procoralan – Servier) is the first in a new class of specific heart rate-reducing drugs and is licensed for the "symptomatic treatment of chronic stable angina pectoris in patients with normal sinus rhythm, who have a contra-indication or intolerance for beta-blockers".7,8 Here we consider the place of ivabradine in the management of patients with stable angina.

Our review of tigecycline (DTB 2008; 46: 62–4) named bacteria that are common causes of skin and soft tissue infection. In doing so, we incorrectly referred to Streptococcus aureus, rather than the intended organism, Staphylococcus aureus (page 62, second paragraph). This error does not affect our article's Conclusion.

Our review of retapamulin (DTB 2008; 46: 76–9) described a trial that compared retapamulin and oral cefalexin in people with infected dermatoses.1 In doing so, we incorrectly stated that the study was underpowered with regards to the number of patients included in its per-protocol analysis. This error does not affect our article's Conclusion.