Medsafe released two alerts regarding A. annua extract in February and November, 2018, detailing 25 reports to the Centre for Adverse Reactions Monitoring (CARM) of liver toxicity, including one case of hepatic cirrhosis. A subsequent investigation of these reports found that the onset of hepatotoxicity occurred within 12 weeks of initiating A. annua extract in most cases, although time of onset varied up to 12 months. Symptoms of hepatotoxicity also varied, but jaundice with pruritis was common among several of the cases. Almost all patients were reported to have recovered or improved following cessation of use, although nine patients required hospital admission. Arthrem has been withdrawn from the Australian market due to safety concerns.

A. annua extract has been promoted as a “natural dietary supplement” for maintaining and supporting joint health and mobility, and it has been used in traditional Chinese medicine to treat fever. Only one small clinical trial has been conducted to date investigating the efficacy of A. annua extract in treating osteoarthritis symptoms, based at the University of Otago and funded by Promisia Ltd (the manufacturer of Arthrem).

• A total of 42 patients were treated over 12-weeks with either placebo (n=14), 150 mg A. annua extract (ART) twice daily (n=14), or 300 mg ART twice daily (n=14)
• Those taking the 150 mg ART dose had a modest and statistically significant improvement in their pain, stiffness and physical function scores
• Patients in the placebo and 300 mg ART groups displayed no improvement; the authors theorised that the sample size of the high dose ART group may have been too small to demonstrate a treatment effect as three patients dropped out, and in addition this group had a higher proportion of patients with severe osteoarthritis than the other two groups

A follow-up at six months reported that the beneficial effects of ART were maintained; however, one patient withdrew from the study due to elevated liver enzymes which the study authors “considered unlikely related to [ART] treatment”

We think the balance of evidence and risk clearly tips towards “no” – and certainly not as a first-line option. There is a very limited evidence base for this treatment, it is not recommended in clinical guidelines for treating osteoarthritis, and as it is an unregulated product the exact composition is uncertain, and includes variable amounts of additional chemicals that have mostly unknown biological/pharmacological functions. Given the strong possibility of hepatotoxicity, prescribers should instead focus on providing evidence-based pharmacological and non-pharmacological interventions.

At the start of the study, the mean baseline weight of the 445 participants was 99.9 kg and mean weight loss after the initial intervention was 8.3 kg; participants attended a 16-week group weight loss programme that included calorie restriction and exercise, and also addressed challenges such as traditional high calorie cooking and lack of exercise facilities in their rural communities. Participants were then randomised into three groups; individual telephone counselling, group telephone counselling and written information only. The telehealth sessions were conducted every two weeks for six months and then monthly for six months. Sessions lasted 10-20 minutes for individuals or 60 minutes for groups and consisted of a review of progress, goal setting, structured problem solving to address any challenges and an action plan for dietary or activity changes until the next session. Participants in the email group received written weight management modules covering the same information. Participants in all groups moved to a self-reliance phase for the final six months, with no contact with health coaches. After a total of 22 months, the mean weight regain was 2.3 kg in the group who received individual telephone support, 2.8 kg for those who received group telephone support, and 4.1 kg for the group who received education material via email. The difference between weight regain in the individual telephone counselling group and the email group was significant, and a larger proportion of those in the telephone group achieved a ≥ 10% weight reduction.

The study authors speculate that the reason the individual counselling group achieved a better maintenance of weight loss was due to the motivating effect of knowing someone would be checking up on them, therefore improving their adherence with the programme, increasing self-monitoring and achievement of calorie and exercise goals. In other words, being accountable for their actions made them more likely to sustain their weight loss. Those in the group sessions achieved better maintenance of weight loss than those who received no counselling, but may have been less comfortable divulging their setbacks and challenges than those who received one-on-one calls, and therefore they did not receive support and problem solving for these individual issues.

A major limitation of applying this model to primary care in New Zealand is the cost involved in conducting individual telephone sessions, to both the patient and the practice, depending on how the sessions were funded. However, it may be a case of balancing the slightly reduced benefits of group counselling, with the lesser cost of this delivery method. At the least, suggesting people instigate a “buddy” system to help motivate them to maintain (or continue) their weight loss may be beneficial.