Intended for healthcare professionals

Clinical Review Lesson of the week

Reye's syndrome and aspirin: lest we forget

BMJ 2001; 322 doi: (Published 30 June 2001) Cite this as: BMJ 2001;322:1591
  1. M C McGovern, specialist registrara,
  2. J F T Glasgow, reader in child healthb (j.glasgow{at},
  3. M C Stewart, senior lecturerb
  1. a Royal Belfast Hospital for Sick Children, Belfast BT12 6BE
  2. b Department of Child Health, Queen's University of Belfast, Belfast BT12 6BE
  1. Correspondence to: J Glasgow, Accident and Emergency Department, Royal Belfast Hospital for Sick Children, Belfast BT12 6BE
  • Accepted 30 August 2000

Aspirin is an avoidable risk factor for Reye's syndrome: heightened vigilance can prevent an increasing incidence

Reye's syndrome represents an abrupt, profound failure of mitochondria, the cause of which is uncertain. It is a biphasic illness, occurring mainly in childhood, which consists of an acute viral prodrome followed several days later by an acute encephalopathy associated with selective hepatic abnormality and metabolic decompensation. A consistent association has been shown between Reye's syndrome and the use of aspirin during the prodromal illness.1 Of the 56 children with Reye's syndrome treated at this centre between January 1979 and December 1986, 46 (82%) had been given aspirin. During the past 13 years just five cases have been seen—two occurred in February 1999, and in both cases the child had been given aspirin.

Case 1

A boy aged 12.5 years had flu-like symptoms of mild fever, headache, and generalised aches for five days. He was given aspirin (300 mg every four hours for 24 hours) after which his symptoms seemed to resolve. However, 12 hours later he began vomiting, and this continued almost hourly for 24 hours, at which time his parents sought medical help.

There was no relevant medical history. The boy was admitted to hospital and managed with intravenous fluids. Twelve hours later he became agitated and uncontrollable; no lateralising neurological features were present and fundoscopy findings were normal. No abnormalities were noted in cerebrospinal fluid or on urine toxicology screening or on a computed tomogram of the brain. The boy's liver transaminase activities and blood ammonia concentration were noticeably high (table). He was admitted to the paediatric intensive care unit for neurological observation and was given intravenous glucose and electrolytes. The confusion resolved within 48 hours, and over the next few days his liver test results became normal. However, the boy remained tired and lethargic and it was three months before he had recovered sufficiently to return to school full time.

Results of investigations in two patients with Reye's syndrome

View this table:

Case 2

A 9 month old boy was seen four weeks after case 1. He had had a low grade fever for about 24 hours, and his mother had given him 150 mg aspirin on one occasion. Twelve hours later he began to vomit and this persisted for 24 hours. When the general practitioner examined him the boy was limp and lifeless, and hypoglycaemia was confirmed (table). The boy improved rapidly after intravenous infusion of a 10% glucose and electrolyte solution, but as he was still drowsy he was transferred to a paediatric intensive care unit. No other neurological features developed. His liver transaminase activities and blood ammonia concentration were raised and the prothrombin time was prolonged (table). Twelve hours after admission to hospital he had a brief generalised seizure; it was not associated with hypoglycaemia, and his cerebrospinal fluid was normal. The seizure responded to intravenous diazepam, and treatment with phenobarbitone was continued for several days. After 24 hours the boy was fully conscious and he subsequently made a full recovery.


The diagnostic criteria of Reye's syndrome were fulfilled in each of these cases. In addition, we excluded inherited metabolic disorders, the group most likely to mimic Reye's syndrome (β oxidation defects, urea cycle disorders, and other organic acidurias2), and other plausible diagnoses.

Reye's syndrome occurred fairly frequently in the late 1970s and early 1980s. In the 24 months between January 1983 and December 1984, there were 26 cases in Northern Ireland. Throughout the 1980s, there was increasing awareness of Reye's syndrome in published reports and the media,3 and of the possible link with aspirin usage.14 Use of aspirin in children under 12 years of age fell sharply after the Committee on Safety of Medicines gave advice to doctors in 1986.5 This fall in usage was documented in Northern Ireland. An audit of febrile children being admitted to hospital in 1988–9 showed a 17-fold reduction in the use of over the counter aspirin when compared with the results of a similar audit carried out in 1985-6.6 On both sides of the Atlantic the number of cases of Reye's syndrome has paralleled this decline in aspirin use. A recent editorial in the New England Journal of Medicine described the trend as “a public health triumph.”7 In 1994-7, only two cases were reported annually in the United Sates,7 and in the United Kingdom and Ireland in 1996–7 there were only five cases.2

We wish to make three points. Firstly, between the end of 1986 and January 1999, only three cases of Reye's syndrome were seen in Northern Ireland in spite of a continuing research interest. In contrast to these three cases, the clinicopathological features collated as the Reye score (table) showed that both cases described here conformed to the characteristics of “classic” (North American) Reye's syndrome—the subgroup of cases shown to be associated with aspirin use.9 In our opinion, cases of this type will continue to occur when the 1986 warning by the Committee on Safety in Medicine is overlooked or disregarded.5 As most parents and many professionals have no direct experience of the severity and possible long term complications of Reye's syndrome, all medicines containing aspirin have, since April 1998, contained an information leaflet stating: “there is a possible association between aspirin and Reye's syndrome when given to children with a fever.” The importance of this advice needs to be more widely appreciated.

Secondly, several effective, alternative treatments (paracetamol, ibuprofen) for symptoms in children with intercurrent infections are not implicated in Reye's syndrome. Children taking aspirin for long term management of connective tissue disorders also run a greater risk of developing Reye's syndrome.10 This was seen in about 4% of 1207 recently reviewed cases in the United States.8 In these circumstances, if a child develops flu-like symptoms or fever, aspirin should be stopped immediately. Although the inherent risks may be greater, similar action might be considered in those requiring aspirin as antiplatelet therapy. In Kawasaki disease, for instance, especially when it is complicated by coronary artery lesions, stopping aspirin prematurely may heighten considerably the chance of thrombosis or even death; specialist advice should be sought and the relative risks carefully balanced. The point is clearly stated in the paediatric formulary, Medicines for Children: “aspirin use should be limited to conditions where a therapeutic advantage is present.”11

Thirdly, the carers of case 1 believed that since his 12th birthday had passed, it was safe to use aspirin for relief of symptoms. Health professionals need to appreciate that this case is not unique. Among 17 cases reported in the United Kingdom since June 1986, 10 were in children aged more than 12 years of age who were given aspirin,12 and in Belay's recent survey of 1207 cases with ages up to 17 years in the United States, 98 (8%) of patients were aged 15 or older.8 The Medicines Control Agency recently decided—incorrectly, we believe—not to extend existing advice on aspirin use to include children aged 12 years and above. If this potentially devastating encephalopathy is not to reappear, there must be heightened vigilance on the part of the public, parents, and health professionals, and continued close monitoring by the Medicines Control Agency.


We thank Professor K Bartlett and Dr M Pourfarzam for carrying out tandem mass spectrometry.

Contributors: All authors were involved in the management of at least one of the patients described; MCMcG collated the clinical details; JFTG initiated the discussion, and MCS and MCMcG contributed to it.


  • Funding JFTG received a research grant (1992–8) from the National Reye's Syndrome Foundation of the United Kingdom.

  • Competing interests None declared.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
View Abstract