There's a lot been written in the blogosphere around what's known as 'complementary & alternative medicine.' (I would argue that there's no such thing - if it works ie improves/cures the patient's health, then it's medicine). In any debate around the use of CAM someone is likely to say that at least it does no harm. For things like homeopathy you could argue that since the client is swallowing only water or sugar pills, with no active principle present, then they're highly unlikely to come to harm (witness the 10-21 homeopathic 'overdose'). The counterargument here is that if the patient relies solely on homeopathy for anything beyond self-limiting conditions then there is in fact considerable potential for harm.
With other 'treatments' the potential for harm is more apparent. And in some cases the harm can be real. In the latest issue of the New Zealand Medical Journal, Brian Kennedy & Lutz Beckert report on the case of a woman whose acupunturist left her with a case of pneumothorax. This is not a trivial problem: pneumothorax is where air builds up within the chest cavity, in the space round a lung, as the result of chest trauma or due to a spontaneous breach in the lung itself - or in this case, because an acupncture needle pierced the lung. This puts pressure on the lung, & as a result the lung collapses. (Pneumothorax has also had medical applications - in Sonja Davies' autobiography, Bread & Roses, she describes it as a treatment for tuberculosis. Apparently collapsing the affected lung makes it more difficult for the tuberculosis bacilli to survive & grow, so the lung has a chance to recover.)
In the case described by Kennedy & Beckert, the patient "became acutely short of breath, following introduction of an acupuncture needle into the right side of her chest posteriorly. She developed 'tightness' ... and associated chest pain" & very sensibly left the clinic, went home, & called an ambulance when her symptoms (typical of pneumothorax) got worse. An X-ray showed that her lung has collapsed, & doctors used a needle to remove 450ml of air from the pleural space around the lung. The next morning the pneumothorax had recurred, which meant surgery to inset a 'drain' into htr chest wall. After the lung reinflated the drain was removed (& presumably the opening was sealed) & she went home a day later.
Madsen, Gotzsche & Hrobjartsson (2009) performed a meta-analysis of clinical trials looking at acupuncture as a treatment for pain. They looked at data from a total of 3025 patients who received either 'real' acupuncture, 'sham' (placebo) acupuncture, & no treatment. Their conclusions: there was "a small analgesic effect of acupuncture ..., which seems to lack clinical relevance and cannot be clearly distinguished from bias. Whether needling at acupuncture points, or at any site, reduces pain independently of the psychological impact of the treatment ritual is unclear." (As Orac comments, on a related study, "the larger and better designed the study, the less likely it is to find a treatment effect greater than placebo due to the treatment.")
Given the following that acupuncture appears to have, people will no doubt continue to seek it out for various ills, regardless of the fact that it performs no better than placebo. In which case, they need to be aware that adverse events like the one described by Kennedy & Beckert, although very rare, can still occur. (These authors list "transmission of diseases, needle fragments left in the body, nerve damage, pneumothorax, pneumoperitoneum [air in the abdominal cavity], organ puncture, cardiac tamponade [accumulation of fluid around the heart] and osteomyelitis [a bone infection]" as major adverse events, albeit extremely rare ones.) They conclude that as these events are generally associated with poorly-trained practitioners, if people do seek out acupuncture treatment they should choose their practitioner carefully - and if treatment involves acupuncture of the chest wall, then the client should be warned about the risks of pneumothorax by the practitioner concerned.
But as Darcy says over on SciBlogs, why go down this route at all?
Brian Kennedy, & Lutz Beckert (2010). A case of acupuncture-induced pneumothorax The New Zealand Medical Journal, 123 (1320) http://www.nzma.org.nz/journal/123-1320/4258
M.V.Madsen, P.C.Gotzsche & A.Hrobjartsson (2009) Acupuncture treatment for pain: systematic review of randomised clinical trials with acupuncture, placebo acupuncture, and no acupuncture groups. BMJ 338: a3115