Pregnant women are one of the most vulnerable groups for developing severe illness related to influenza AH1N1 infection. During the first outbreak from April to June 2009, the Centers for Disease Control and Prevention of the USA reported six (12%) maternal deaths among 49 total deaths with a confirmed diagnosis of influenza AH1N11. In Mexico, the prevalence of maternal deaths among infected pregnant women varied from 7–10%, and these were mainly associated with acute respiratory distress syndrome (ARDS) and multi-organ failure2. In affected women the reported perinatal morbidity included fetal and neonatal death, intrauterine growth restriction and preterm delivery. Despite the increased prevalence of perinatal complications in these patients, no reports on the process of fetal hemodynamic adaptation/deterioration during maternal influenza AH1N1 infection are available. We report here the fetal Doppler changes observed in pregnant women during the acute phase of influenza AH1N1 infection and their association with maternal peripheral blood oxygen data and with the incidence of perinatal complications. Fifteen pregnant women with acute clinical symptoms of lung infection and a diagnosis of influenza AH1N1 confirmed by real-time reverse-transcription polymerase chain reaction were evaluated. Four patients had additional morbidity: asthma (n = 1), hypothyroidism (n = 2) and systemic lupus erythematosus (n = 1). In six patients, a superimposed bacterial pneumonia was diagnosed based on clinical symptoms and on visualization of areas of lung consolidation on thoracic radiogram. All women received antiviral treatment (oseltamivir) and antibiotics (ceftriaxone, clarithromycin or ampicillin). Blood gases on admission showed five patients with hypoxemia (defined as partial pressure of oxygen (PaO2) < 60 mmHg), and 10 patients with normal blood gases. Respiratory support was provided by a mixture of air and oxygen (oxygen concentration of 24%), at 5 L/min until respiratory symptoms improved. Blood gas values were evaluated every 6 h during the first day, and then repeated as required. Patients with hypoxia increased their PaO2 from < 60 mmHg to > 80 mmHg within the first 6 h after admission. Clinical symptoms improved on average after 2.5 (range, 2–4) days from starting treatment, and respiratory symptoms improved on average after 3.6 (range, 2–5) days. For analysis the patients were divided into two groups: (1) those presenting with blood gas values indicative of hypoxemia (PaO2 < 60 mmHg) and those with normal blood gas values (PaO2 ≥ 60 mmHg). Ultrasound and Doppler examinations were performed within 24 h from the time of admission, 3 days after admission and 4 weeks after admission (Table 1). Fetal Doppler evaluation showed that fetuses from women with hypoxia had significantly higher Z-scores in the umbilical artery pulsatility index (PI), reduced Z-scores in the middle cerebral artery PI (MCA-PI) and increased myocardial performance index within 24 h of admission than did fetuses from women with normal blood gas values; the ductus venosus data did not show any differences among the groups. Doppler parameters did not exceed normal reference values for gestational age, and no instances of absent or reversed diastolic flow in the umbilical artery or absent atrial flow in the ductus venosus were registered. After 3 days from admission, the patients had an improvement in clinical symptoms but the fetal Doppler findings were similar to those observed at admission (Figure 1). Four weeks later, only the MCA-PI was different between the two groups. Maternal complications included one patient with uterine contractions at admission, which were controlled with intravenous fluids, and six patients with long-term complications, comprising all five patients from the hypoxia group (one case of preterm labor, two preterm deliveries, one at < 34, and one at > 34 weeks' gestation, and two cases of fetal growth restriction), and one patient with fetal growth restriction from the non-hypoxic group (P < 0.01). The results of this study show that among pregnant women with AH1N1 influenza, those with reduced PaO2 had significant differences in fetal hemodynamic parameters from those with normal blood gas values. These differences were still evident 3 days after starting treatment, despite improvement in the maternal clinical condition. Four weeks after the initial scan, no differences in Doppler parameters were observed between the two groups except for MCA-PI. Influenza AH1N1 infection increases alveolar secretion and produces a local and systemic inflammatory response and an increase in extravascular lung water3. The ratio between PaO2 and the fraction of inspired oxygen (FiO2) can be altered, leading to ARDS. Reduced maternal PaO2 might provoke an acute hypoxic insult to the placenta and fetus, precipitating a process of fetal blood centralization characterized by increased vascular resistance in the umbilical artery and reduced resistance in the fetal brain vessels. The fetal heart also shows initial signs of adaptation by increasing the myocardial performance index. According to our results this process seems to be effective, at least in the acute phase of respiratory complications, as Doppler values did not exceed normal ranges for gestational age, and the ductus venosus was not altered in the hypoxic group. The adaptive response was still observed 3 days later despite improvement in maternal respiratory symptoms. Four weeks later, three of the four Doppler parameters we tested were similar among the two groups; only the MCA-PI stayed at a reduced level in the hypoxic group. Despite this hemodynamic adaptation, the prevalence of short- and long-term complications was high (40%) and similar to that previously reported, with hypoxic women showing the highest prevalence of perinatal complications4. The general recommendation in the USA is that all pregnant women should receive immunization to protect against influenza AH1N1 infection, but nearly 50% of them do not receive the vaccine5. More information needs to be provided to pregnant women on the benefits of receiving immune protection. Otherwise, more and increasingly severe cases of influenza AH1N1 will continue to be reported every year. Our results suggest that close fetal surveillance during the acute episode of respiratory symptoms and after recovery in women with influenza AH1N1 infection can identify fetuses at a higher risk of abnormal perinatal outcome. This study was supported by the Mexican National Council of Research and Technology (Consejo Nacional de Ciencia y Tecnologia, CONACyT) project: SALUD-2009-C02-127102. E. Hernandez-Andrade*†‡, R. Figueroa†, A. Cerbulo-Vazquez†, J. A. Benavides-Serralde†§, G. P. Borbón† and J. M. Ramírez† †National Institute of Perinatology, Mexico City, Mexico; ‡Department of Obstetrics and Gynecology, Division of Maternal-Fetal Medicine, Wayne State University, 3990 John R, 4 Brush, Detroit, MI, 48201 USA; §Department of Obstetrics and Gynecology, Technological University of Pereira, Pereira, Colombia *Correspondence. (e-mail: [email protected])