Recent findings about the clinical consequences of the 1918 pandemic – two further studies describing male autopsy results and female birth rate decline and miscarriagesArchived

ECDC comment

The 1918 pandemic continues to provide a rich source of studies of the clinical impact of those novel viruses which between 1918 and 1920 killed up to 50 million people world-wide. These two recent studies first shows autopsy results among military recruits who died from the first influenza pandemic of the 20th century and the second suggests the impact on births.

The 1918 pandemic continues to provide a rich source of studies of the clinical impact of those novel viruses which between 1918 and 1920 killed up to 50 million people world-wide. These two recent studies first shows autopsy results among military recruits who died from the first influenza pandemic of the 20th century and the second suggests the impact on births.

Autopsy series of 68 cases dying before and during the 1918 influenza pandemic peak Zong-Mei Sheng ZM, Chertow DS, Ambroggio X, et al. PNAS 2011; published ahead of print September 19, 2011

This article describes the clinical, pathological, bacteriological and some virological findings in the preserved lung tissues of 68 fatal influenza/pneumonia male military patients dying between May and October of 1918 in the USA. This is a period that spans from 4 months before the 1918 pandemic was recognized to 2 months during the pandemic appeared and rapidly peaked. The pre-pandemic and pandemic peak cases were indistinguishable pathologically in the cases. All 68 cases had histological evidence of bacterial pneumonia and 94% showed abundant Gram positive bacteria. The lung tissues of only 37 of the 68 cases were positive for influenza viral antigens or viral RNA, including four from the pre-pandemic period (May–August). Sequence analysis of the viral hemagglutinin receptor-binding domain performed on RNA from 13 cases are suggestive of a trend of viral evolution from a more ‘avian-like’ viral receptor specificity with G222 in the pre-pandemic cases to a more ‘human-like’ specificity associated with D222 in pandemic peak cases. However the numbers are small and the authors also note that viral antigen distribution in the respiratory tree was no different between pre-pandemic and pandemic peak cases, or between infections with viruses bearing different receptor-binding polymorphisms. The 1918 pandemic virus circulated for at least 4 months in the United States before it was recognized epidemiologically in September 1918. The authors conclude that the causes of the unusually high mortality in the 1918 pandemic were not explained by any apparent pathological and virological parameters examined in their study.

Natality Decline and Miscarriages Associated With the 1918 Influenza Pandemic: The Scandinavian and United States Experiences Bloom-Feshbach K, Simonsen L, Viboud C, et al. J Infect Dis. (2011) 204(8): 1157-1164

The authors of this epidemiological study examined the relationship between the 1918 pandemic and trends in birth rates during the 1918 pandemic in three Scandinavian countries (Denmark, Sweden and Norway) and the United States with the purpose of elucidating a possible effect of this pandemic influenza on miscarriages and births. The methodology they used was based on a compilation of monthly birth rates from 1911 through to 1930 in the four countries, as well as on identified periods of unusually low or high birth rates. With these data they quantified births as ‘missing’ or ‘in excess’ of the normal expectation. Using monthly reports of influenza in the country they looked for correlations in the timing of peak pandemic exposure and depressions in birth rates, and identified pregnancy stages at risk of influenza-related pregnancy losses. They found that uniquely in the nineteen year period birth rates declined in all study populations in the spring of 1919 by a mean of 2.2 births per 1000 people. This represents a significant fall of between 5% and 15% below baseline levels. The depression in birth rates observed in 1919 reached a trough 6.1 to 6.8 months after the autumn pandemic peak. This is compatible with the missing births being attributable to excess first trimester miscarriages in approximately 1 in 10 women who were pregnant during the peak of the 1918/1919 influenza pandemic. The authors examined a number of potential competing explanations. They found that pandemic-related female mortality could not explain the observed patterns. Moreover, they suggest causality on grounds of the temporal synchrony across geographical areas. After the birth decline the authors found a compensatory increase in births 7.5 months to ten months later. In Denmark it was possible to specifically study the trends in still-births and distinct trends in rural Denmark and urban Copenhagen. The authors found that there was also an excess of stillbirths in that country in association with the pandemic wave. They also found that the trough of missing births occurred two months earlier in Copenhagen than in the rural areas which paralleled a two month delay in the A(H1N1) 1918 pandemic.  

ECDC Comment (10th October 2011):

The male autopsy article continues an ongoing debate concerning why the second autumn 1918 pandemic wave was considerably worse in many countries than the first wave spring wave in causing deaths. The hint in this paper is that the virus evolved to something more adapted to humans and more pathogenic between the spring and autumn waves of the 1918 pandemic. However there is at least one competing hypothesis, namely that the first wave of infection provided some cross-protection against a more intense second wave.  The study by  Zong-Mei Sheng really cannot resolve this.  

The miscarriage and birth trends study is more intriguing as it is looking at a subtle potential impact of influenza (3). Usually, miscarriages are not recorded in vital registration or epidemiological studies. In retrospective studies like this all that can be done is to infer such effects by noting declines in birth rates and perhaps compensatory rises in births in the period after as is recoded here. It is important to note that there was also an association with still-births recorded in the 2009 pandemic in the European study with the most complete follow-up. An accompanying editorial tries to stretch the 1918 findings with others to encourage vaccination of all pregnant women against seasonal influenza. This seems rather strong given the principal findings are from the pandemic of 1918. Really there is a need now in Europe for close studies of the impact of the new seasonal influenza on all aspects of pregnancy, including miscarriages.