An article published Online First and in a future edition of The Lancet reports that a study of Mexican swine flu cases shows that infants and people aged 39 years and under are the most likely to get infected. However, elderly people have the highest swine flu mortality rates. The data covers the period up to July 31, 2009. The article is the work of Dr Victor Borja-Aburto, Mexican Institute for Social Security, Mexico City, Mexico, and colleagues.

The influenza surveillance system gathered information from April 28 to July 31, 2009 on patients with influenza-like illness who attended clinics that were part of the Mexican Institute for Social Security network. The authors analysed the reports. Findings indicate that by July 31, 63,479 cases of influenza-like illness were reported; 6,945 (11 percent) cases of H1N1 were confirmed. Of these, 6,407 (92 percent) were outpatients, 475 (7 percent) were admitted and survived, and 63 (less than1 percent) died. Individuals aged 10 to 39 years were the most affected with 3,922 cases (56 percent). The following mortality rates were recorded in other age groups:

• 60-69 years 5.7 percent;
• 50-59 years 4.5 percent;
• 40-49 years 2.7 percent;
• 30-39 years 2.0 percent;
• under 1 year 1.6 percent;
• 20-29 years 0.9 percent;
• 1-9 years 0.3 percent;
• 10-19 years 0.2 percent.

Mortality rates showed a J-shaped curve, with greatest risk in those aged 70 years and older (10.3 percent).

In individuals who had been vaccinated for seasonal influenza, risk of infection was lowered by 35 percent. After the fourth day of symptoms onset, every day of delay in hospital admission increased the risk of death by almost 20 percent. In addition, presence of chronic diseases in a person increased their risk of death six-fold.

The authors mention that pregnant women accounted for 6 percent of the deaths in their study (four of 63). This is slightly lower than the rate recorded in the USA during a similar period (8 percent, seven of 87 deaths). The authors say: "In Mexico, all pregnant workers were sent home during the peak of the pandemic, which probably accounts for this difference. Information about pregnancy in ambulatory confirmed cases was not available in the surveillance system, and we were unable to estimate the case fatality rate for pregnant women. None of the pregnant patients who died received antiviral drugs during the first 48 h and none had received influenza vaccine. Treatment should begin immediately after onset of symptoms in this group, and vaccination during pregnancy is not contraindicated and therefore can be considered."

The authors acknowledge that possible protection against H1N1 flu through the seasonal influenza vaccine is controversial. They explain: "The Mexican population who has received seasonal influenza vaccination since 1977, including H1N1 components, could have benefited from cross immunity. The high incidence of infection in young people could show not only their different exposure related to their daily activities but also that people older than 60 years might have some immunity against the H1N1 virus."

The authors write in conclusion: "Although H1N1 has spread to 168 countries and territories worldwide, it has not reached the dimensions of its great predecessor in 1918, and some researchers believe, with the information available up to now, that the present H1N1 influenza virus will not cause a pandemic on the scale of those during the 20th century. This pandemic might not be the one we expected; however, the virus is evolving and the threat continues."

In an associated comment, Dr V Alberto Laguna-Torres, San Marcos University, Lima-Peru, and, US Naval Medical Research Center Detachment, Lima, Peru and Dr Jorge Gomez Benavides, Ministry of Health of Peru, Lima, Peru, remark: "A significant finding in today's report was the higher infection risk in children, but the disease was more severe (ie, higher risk of death) in people older than 60 years. Additionally, school closures and restrictions of mass gatherings to mitigate the spread of the epidemic was successful, as has been seen in other countries."

In closing, they say: "Decisions based on preliminary results and limited sources have to be made, and sometimes there is no time to wait for the pandemic to end to have stronger information. Currently we know that the pandemic has not reached the dimensions of its predecessor in 1918, but the scientific knowledge has evolved faster than before, probably because of global and online communication."

"Infection and death from influenza A H1N1 virus in Mexico: a retrospective analysis"
Santiago Echevarría-Zuno, Juan Manuel Mejía-Aranguré, Alvaro J Mar-Obeso, Concepción Grajales-Muñiz, Eduardo Robles-Pérez, Margot González-León, Manuel Carlos Ortega-Alvarez, Cesar Gonzalez-Bonilla, Ramón Alberto Rascón-Pacheco, Víctor Hugo Borja-Aburto
DOI: 10.1016/S0140-6736(09)61638-X
The Lancet

Stephanie Brunner (B.A.)