Sudden acute respiratory syndrome

Plagues are as certain as death and taxes.¹ The optimism of the 1960s and 1970s has given way to a mature realism that the relationship between human beings and microbes is neither completely predictable nor biased in favour of humans. Over the past few decades several important human viruses have emerged. Some, such as HIV, prove to be sustainable modern plagues adding to the toll of human misery. Others, such as hepatitis F, occupy a seemingly silent niche, passengers in a human caravan but contributing little to the joint relationship. Whereas viruses such as Ebola, Hantaan, and Nipah spring from an animal reservoir, destroying life but unable to sustain transmission in a new environment, others such as human metapneumovirus² are associated with respiratory illness in young children but their contribution to adult disease remains uncertain, suggesting a balance between virus and host immune system achieved after some evolutionary negotiation. Each of these viruses has been identified through the study of human disease processes, each of which exists along a spectrum of possible outcomes between virus and host.

Emergence of new diseases

Against this background, the emergence of new human infectious diseases or viruses is unsurprising. Severe acute respiratory syndrome was first recognised at the end of February in Hanoi, Vietnam.³

The agent is highly infectious, with attack rates of >50% among healthcare workers caring for patients with the syndrome.⁴ Preliminary data from the first cluster of about 60 probable cases in Hanoi indicate an incubation period of 5-9 days. The most common early systemic symptoms in Hong Kong and Hanoi include fever, malaise, myalgia, headache, and dizziness. Sore throat and rhinorrhoea occur early in fewer than 25% of cases, and cough occurred early in only 39% of cases.⁴ Because of its non-specific early manifestation, sudden acute respiratory syndrome will be overlooked unless clinicians have a high index of suspicion and seek a history of travel or contact with the syndrome.

After 3-7 days of fever the lower respiratory phase begins, with a non-productive cough, which may be accompanied by dyspnoea and chest pain.⁵ Breathlessness requiring oxygen occurred in many cases after about five days and progressed to hypoxaemia requiring ventilatory support in around 15%, a rate similar to the 10-20% observed elsewhere.⁵ Early chest x ray findings typically show small focal unilateral diffuse interstitial infiltrates, which may be overlooked initially. The appearance evolves rapidly, often becoming more generalised and affecting both lung fields. Chest radiographs may, however, be normal during the febrile prodrome and throughout the illness.⁴ Lymphocytopenia is common and occasionally liver function values are raised.⁴

Clinical presentation suggests an illness of variable severity ranging from mild illness to death. The speculation is that the most severe illnesses occur among first level contacts of an index case. If real, this may reflect either repeated high dose exposure of the unsuspecting healthcare workers to the index case or attenuation of the pathogen during subsequent waves of infection.

Management

The influenza neuraminidase inhibitor oseltamivir and antibiotics targeted at known bacterial pathogens causing atypical pneumonia have been used without evident benefit. Treatment in several places has included steroids and the antiviral agent ribavirin administered intravenously. Their efficacy remains unproved, but they may have been responsible for some clinical improvement seen in critically ill patients in Hong Kong.⁵ Further evaluation of ribavirin is urgently required, both in cases and the laboratory, particularly as intravenous ribavirin is expensive and not widely available. Fortunately, many cases of probable sudden acute respiratory syndrome improve steadily over 7-10 days without complications or a need for supplemental oxygen. The case fatality rate among cases meeting the current WHO case definition is about 3%.

This outbreak raises several important clinical issues: how to diagnose cases rapidly; the response to antivirals; the duration of virus shedding (which affects the timing of discharge and return to work); the presence and duration of viraemia; the distribution of the virus in nature; and whether the pathogen is highly variable. In Hanoi, as in other parts of the world, the brunt of the outbreak has been borne mostly by healthcare workers having direct contact with cases. Other patients and family have been affected to a lesser extent. These observations indicate that the syndrome is transmitted by droplets, but transmission in some cases remains unexplained. Until the route(s) has been clearly established infection control measures should include both airborne precautions (including a negative pressure isolation room, use of full respiratory protection for people entering the room, and eye protection for all contacts) and contact precautions (gowns and gloves and hand hygiene).

By the third week in March several hundred probable cases of the syndrome had been reported worldwide, with epidemiologically linked clusters in Hanoi, Hong Kong, Singapore, and Toronto, and further linked cases in New Jersey, California, and Bangkok. There is press speculation about a link between the clusters to the ninth floor of a hotel in Hong Kong, where a doctor from Guangzhou, Guandong Province, China, who had been exposed to patients with the syndrome in Guandong, and nine other cases were staying. Thus the current global outbreak may have evolved from an outbreak of a similar respiratory condition in Guandong last November.⁶ The means of transmission in the hotel is under investigation: droplet spread in the lift lobby is the most likely.

The search for the cause

The speed of travel favours intercontinental spread of disease. The rapid dissemination of sudden acute respiratory syndrome around the world should be considered a rehearsal for the next pandemic of influenza,⁷ as it shows what will happen with a new human virus spread by the respiratory route, with no vaccines and antivirals in limited supply. However, the speed of communication in the virtual world is an advantage to the microbial detective. The tried and trusted forensic approaches of the classical virologist, the electron microscope and the tissue culture plate, become powerful investigative tools when the images of a suspect can be shared immediately between laboratories thousands of miles apart. When these approaches are combined with real time polymerase chain reactions, differential display technology, and generic molecular identity tests designed to catch all viruses in particular families, the rate of data development is exponential. It is possible to undertake a microbial identity parade and go from patient sample to microbial nucleic acid detection, sequence analysis, and phylogenetic tree characterisation, in less than 12 hours—if you know what to look for. Nevertheless, it may still take weeks or months to catch the culprit in a new disease. So far among the candidates a leading contender seems to be a paramyxovirus. However, there is no substitute for sifting scientific evidence carefully and slowly assembling fragmentary pieces of the puzzle to provide a complete picture and a testable theory of causality, which is all the more convincing when it can be tested simultaneously in several laboratories using material from many different patients.

The advantages of real time communication are also exploited by the media, who can track the progress of the disease and profile afflicted individuals, put the spotlight on affected institutions, and seek accountability from those trying to contain the impact of new diseases. The techniques of tracking a new disease parallel those of tracking a war and involve documenting death and detritus, progressing up blind alleys, reporting spectacular highlights, and asking unanswerable questions, emphasising that emerging infectious diseases and mortal combat may still have much in common. Our mastery of the microbial world is less complete than we might imagine and more subject to chance interactions in the environment than we might care to admit.

News p 677