Pressure is mounting on a Beijing neurosurgeon to prove that his popular treatment for spinal-cord injury works. In an article published last month, a group of spinal experts concludes that the treatment, which involves implanting fetal cells into the spine to promote nerve-cell regeneration, has significant side effects and does not provide any benefit.

Hongyun Huang's technique is based on the theory that olfactory ensheathing cells (OECs), which normally help to link nerve cells in the nose and the brain, can help regenerate nerve cells at the site of an injury. Since 2001, Huang, who works at Chaoyang Hospital, has treated around 600 patients with tissue from aborted fetuses that he says contain OECs (see Nature 437, 810–811; 2005).

Three spinal-cord experts have now published a critique of his methods (B. H. Dobkin, A. Curt and J. Guest Neurorehabil. Neural Repair 20, 5–13; 2006); the researchers followed seven of Huang's patients before and after treatment, reviewed his publications and visited his lab.

They say Huang's surgical techniques are good. But despite Huang reporting at a February 2004 meeting that there had been no adverse effects of more than 500 implantations, the researchers found that five of the seven patients experienced side effects including meningitis. They also question whether the cells used by Huang are OECs. “We don't know what those cells are but they are not pure OECs,” says first author Bruce Dobkin, a neurorehabilitation specialist at the University of California, Los Angeles.

There is a good chance patients are just wasting money and expectations.

The most damning claim in the critique is that none of the patients showed any improvement after treatment. For example co-author Armin Curt, a neurologist at the International Collaboration on Repair Discoveries (ICORD) at the University of British Columbia, Canada, measured muscle responses to nerve signals, and found no change in the three patients he studied before and after the procedure. “There is a good chance patients are just wasting money and expectations,” says Curt.

Huang vigorously defends his technique and calls the report a “pack of lies”, pointing out that the third author, James Guest of the Miami Project to Cure Paralysis, has previously reported observations that support his procedure (J. Guest, L. P. Herrera and T. Qian Spinal Cord 44, 135–142; 2006). Several patients contacted by Nature claim that they have experienced benefits. A nurse practitioner in California says her son showed various improvements, such as reduced sweating, muscle spasm and pain, and a better sense of balance since he had the surgery in November 2004, and accuses Dobkin of focusing on particularly difficult cases.

But Dobkin says such testimony can be misleading, because patients are so desperate to believe that they have improved after forking out US$20,000 (or $3,700 for Chinese patients) and undergoing a risky procedure. “They forget what they were like before,” he says. The surgery itself could also lead to short-lived improvements by relieving pressure in the area, adds John Steeves, a spinal-cord injury specialist at ICORD. “You really need long-term follow-up,” he says. Steeves, who has visited Huang's lab, says Huang has consistently ignored advice on how to use blind assessment, randomized controls and long-term observation.

Other spinal-cord treatments offered in countries such as Brazil and Portugal are also gaining popularity despite the lack of such data. This has prompted a group of international researchers to draft guidelines to help patients and clinicians evaluate treatments. Sponsored by the International Campaign for Cures of Spinal Cord Injury Paralysis, the guidelines include recommendations and hard data to help distinguish the effects of a given treatment from other factors, says contributor Mark Tuszynski, a neuroscientist at the University of California, San Diego. For example, 40% of individuals show some spontaneous recovery after acute spinal-cord injury, so anecdotal stories of improvement don't necessarily mean a treatment works. The guidelines will be submitted to an academic journal this summer, and a simplified version translated into many languages.

But getting the word out may not be easy. Paul Lu is a postdoctoral student in Tuszynski's lab who entered neurology after he was paralysed from the waist down in a car accident. A Chinese native, Lu has been trying to warn patients in China about Huang's procedures, but says his translation of the Neurorehabilitation and Neural Repair article was rejected by several journals and science-based newspapers. “Chinese journals like to claim that China is leading the world in cell transplants,” says Lu. “They'd lose face if they print this.”

Last week, however, the Chinese Journal of Spine and Spinal Cord agreed to publish the translation in June. Readers will have to balance such reports with Huang's confidence, buoyed by patient testimony, in his therapy.