Zika virus infection is usually asymptomatic or mild. However, it can sometimes lead to an autoimmune disease called Guillain-Barré syndrome. In rare patients, this can result in serious neuropathic complications that require urgent hospital treatment. During an outbreak health authorities must provide enough beds in intensive care to treat these patients. This was the case in Martinique, a year ago, during the 2015-2016 Zika outbreak that hit the entire American continent. Santé Publique France, who provided recommendations to Martinique, relied on the modeling work of Simon Cauchemez’s team at the Institut Pasteur to analyze the situation.
Detected in north-western Brazil in May 2015, the Zika virus has spread very quickly and Brazil has reported the highest ever number of cases to date with between 440,000 and 1,500,000 suspected cases reported. The virus has gradually spread over the American continent. In December 2015, the first two cases were detected in French Guiana by the Institut Pasteur in French Guiana, followed by two other cases in Martinique.
The winter of 2015-2016 “was the start of the Zika outbreak in Martinique and we wondered whether we’d have enough respirators on the island to care for any patients with Guillain-Barré syndrome”, explains Harold Noël, a physician-epidemiologist at Santé Publique France (formerly the Institute for Public Health Surveillance, InVS). In March 2016, researchers from the Institut Pasteur, and local partners in French Polynesia, showed that, although often asymptomatic or mild, the Zika virus infection is associated with the development of Guillain-Barré syndrome. This rare autoimmune disease sometimes leads to paralysis that can affect the respiratory muscles.
But how can you anticipate hospital care for seriously ill patients? “In Martinique, not only did we have to find enough respirators, we also had to call in additional qualified staff from mainland France,” continues Harold Noël.
To improve its predictions, Santé Publique France called for technical support from the Institut Pasteur in Paris and its Mathematical Modeling of Infectious Diseases Unit: “We analyzed the data available to us in Martinique regarding Zika virus monitoring and Guillain-Barré syndrome”, explains Simon Cauchemez, who leads the unit. “In particular, we developed methods for analyzing the progression of the outbreak in real time, week after week.” The data was updated weekly, enabling the modeling team to improve its predictions. After an initial model training phase, “our findings indicated that the number of cases of Guillain-Barré syndrome were going to be significantly lower in Martinique than was expected from a simple extrapolation of the outbreak observed in French Polynesia two years earlier.” The researchers estimated that eight intensive care beds and seven respirators would be sufficient to deal with the situation and thus confirmed that the resources made available by the French authorities were adequate (see Lazarus C, Guichard M, Philippe JM, et al. The French experience of the threat posed by Zika virus. Lancet, July 2016).
“This modeling was a complex task,” points out Simon Cauchemez, “because we didn’t know what proportion of infected people actually went to see a doctor. It was therefore difficult for us to extrapolate the number of persons infected on the island.” The team therefore tested several possible scenarios, especially by comparing data from the 2013-2014 Zika outbreak in French Polynesia.
“When we monitor an emerging disease, we have to accept that we don’t always have all the data and this makes analysis difficult,” continues Simon Cauchemez. “Modeling work often involves ascertaining the degree to which something we don’t know affects the predictions.” It also involves educating the health authorities. Modeling does not predict the precise behavior of an outbreak but it does provide working hypotheses to anticipate several possible scenarios. “Understanding the relevance of modeling is essential for making informed decisions”, explains the Santé Publique France doctor. “We have therefore, in turn, been able to educate the General Directorate of Health and the Regional Health Agency.”
This research showcases the contribution of modeling to local health care planning during outbreaks. Further studies of this type are needed to improve on the predictive power of such approaches.
Real-time assessment of healthcare requirements during the Zika virus epidemic in Martinique, American Journal of Epidemiology, February 15, 2017.
Alessio Andronico, Frédérique Dorléans, Jean-Louis Fergé, Henrik Salje, Frédéric Ghawché, Aissatou Signate, Elise Daudens-Vaysse, Laure Baudouin, Timothée Dub, Maite Aubry, Van-Mai Cao-Lormeau, Martine Ledrans, Harold Noel, Henri-Pierre Mallet, Arnaud Fontanet, André Cabié, Simon Cauchemez
Author affiliations: Mathematical Modeling of Infectious Diseases, Institut Pasteur, Paris, France (Alessio Andronico, Henrik Salje, Simon Cauchemez); Santé Publique France, French National Public Health Agency, Antilles-French Guiana Regional Intervention Unit (Cire), Saint-Maurice, France (Frédérique Dorléans, Elise Daudens-Vaysse, Martine Ledrans); Martinique University Hospital, Fort de France, France (Jean-Louis Fergé, Aissatou Signate, André Cabié); Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, USA (Henrik Salje); French Polynesia Hospital Center, Papeete, Tahiti, French Polynesia (Frédéric Ghawché, Laure Baudouin); Emerging Diseases Epidemiology Unit, Institut Pasteur, Paris, France (Timothée Dub, Arnaud Fontanet); Institut Louis Malardé, Papeete, Tahiti, French Polynesia (Maite Aubry, Van-Mai Cao-Lormeau); Santé Publique France, French National Public Health Agency, Saint-Maurice, France (Harold Noel); Public Health Surveillance Office, Directorate of Health, Papeete, Tahiti, French Polynesia (Henri-Pierre Mallet).