• severe acute respiratory syndrome (SARS)

In February 2003 the World Health Organisation (WHO) received alarming news of antibiotic resistant community acquired pneumonia in Vietnam, Hong Kong, and Singapore of apparent viral origin. It named the disease “severe acute respiratory syndrome” (SARS) and issued urgent advice directed at reducing transmission and spread (http://www.who.int/csr/media/sars_wha.pdf). With unprecedented speed, the probable cause was identified as a novel coronavirus, now named SARS CoV. The imposition of draconian public health measures appears to have brought the disease under control, but there is still concern that, with over 8000 suspected or confirmed cases and contact networks reaching millions, there is every prospect that the disease will become endemic in China or spread to areas of the world in which it cannot be contained.

Although co-infection with SARS CoV and another agent has not been ruled out as the cause of SARS (particularly in “superspreaders” causing numerous secondary cases), SARS CoV alone seems likely to be the cause of most cases of SARS and can reproduce the disease in non-human primates.^1,2 Coronaviruses have the distinction of containing the largest genome of all known RNA viruses. They are widespread throughout the animal kingdom, causing bronchitis (poultry), hepatitis (mice), enteritis (horses and pigs) and peritonitis (cats), but often infecting multiple sites. Known coronaviruses fall into three clades (members of two of which cause about 20% of common colds), but the SARS CoV occupies a new fourth clade of its own. It is clearly a coronavirus, but not closely related to any previously sequenced virus. It presumably arose from an animal source in southern China, perhaps in a species with relatively little contact with man and in which viral disease has been little studied. The civet cat has been proposed as a possible source, but systematic studies of coronaviruses in a wide spectrum of wild and semi-domestic species are not yet complete. RNA viruses tend to evolve rapidly and coronaviruses frequently undergo homologous recombination, so that co-infection with an established human coronavirus and SARS CoV could lead to emergence of new virus species combining various features of the parental strains.

The clinical picture of SARS CoV infection continues to emerge, but patients in the early stages may not complain of respiratory symptoms and my not be febrile; influenza-like symptoms, abdominal pain, and diarrhoea are common,³ followed by transient high fever and then by ARDS, progressive respiratory failure, spontaneous pneumomediastinum, and disseminated intravascular coagulopathy (DIC). Full recovery may take weeks or months.⁴ The WHO definitions were established to assist in the definition of hospital cases and have a sensitivity of only 26% in the detection of non-hospitalised patients defined by seroconversion.³ Particular problems with SARS CoV include:

• the diagnosis is difficult;

• it spreads fast in hospital wards (especially if nebulisers are used or emergency resuscitation is performed);

• even fit, healthy people (including hospital workers) are affected; and

• there are no proven specific treatments.

SARS has knock-on effects on the care of other patients and disrupts the lives of relatives and hospital staff alike. A major SARS outbreak in the UK could effectively close down the health service.

In this issue of Thorax, Chan et al⁵ describe the clinical features of 115 patients (including five doctors and 18 nurses) with SARS admitted to a single hospital in Hong Kong, beginning in March 2003. Their mean age was 41 years, and the crude mortality was 15.7% with one third of deaths occurring more than 3 weeks after onset of symptoms. Intriguingly, Chan et al show that diabetes, cardiac disease, and age are strongly predictive of an adverse outcome, mirroring a smaller previous study of patients mainly from the Amoy Gardens housing block. In this other study⁴ 40% of those with ARDS (n=15) had chronic hepatitis B infection compared with 5% of the 60 patients who did not develop ARDS.

These observations raise intriguing questions about why some patients become very ill and die while others have mild disease and survive. Certainly, young healthy people infected with SARS CoV rarely become very ill,⁶ and those already elderly⁷ or in poor health are at increased risk when a viral or bacterial pneumonia develops. However, it is also possible that some of the patients with pre-existing disease may have heightened innate immune responses that augment the immunopathological response to SARS and thus leads to more severe pulmonary infiltration, ARDS, and death. It is also possible that the immunological traits that lead to chronic disease—for example, hepatitis B infection or immune organ damage—also adversely influence the immunopathological response to SARS CoV infection.

Experience with smallpox and polio shows that a highly effective vaccine is essential for global elimination of an infectious disease and that an animal reservoir makes elimination hard or impossible. Vaccine development is a worldwide priority, funded by US Federal support for industrial partners using three distinct approaches. A vaccine would be likely to prevent systemic spread; there are successful vaccines for some veterinary coronavirus infections and it would be possible to test vaccines in non-human primates.² However, success is not guaranteed and anti-coronavirus immunity can even increase disease severity—for example, in coronavirus induced feline peritonitis. The existing human coronavirus common cold agents are able to re-infect despite low variability, and prolonged viral shedding in SARS patients (about 70% of patients are still positive at day 21 on stool samples) despite good serological responses (60% seroconversion by day 21 and virtually 100% by day 30⁴) indicates that a specific immune response may not be capable of terminating infection.

The SARS outbreak has important lessons for us all. Epidemics of this type do not respect national borders, have a large impact on tourism, travel and trade, and potentially have devastating effects in poor countries with insufficient infrastructure. The unprecedented speed of international and national collaboration undoubtedly contributed greatly to limiting the impact of SARS, and the WHO and respective governments must be praised for their incisive and energetic leadership. What will happen to SARS during the next 6–9 months is guesswork—a major worldwide epidemic might develop this coming winter or the outbreak could die down. Certainly, there will be more outbreaks of respiratory viral disease in the future, and we need to be well prepared for such events.