Date Published: June 14, 2019
Publisher: Public Library of Science
Author(s): Olivia Anselem, Camille Baraud, Anne-Sophie L’Honneur, Camille Gobeaux, Flore Rozenberg, François Goffinet, David J. Garry.
During winter, after excluding obvious sites of infection, the most important diagnoses of isolated fever or influenza-like illness (ILI) to rule out are listeriosis and influenza, because of their severe potential outcomes and the straightforward management available for each. While awaiting laboratory results, the recommended management strategy is usually hospitalization for intravenous antibiotic therapy against potential listeria. This study sought to assess the effect of the use of a rapid test on hospitalization and antibiotic therapy rates.
The study included all pregnant women who consulted for ILI or isolated fever after clinical and laboratory investigations and had a molecular diagnostic assay for influenza during two time periods, both during influenza epidemics: before introduction of the rapid molecular assay use (period 1) and after this (period 2).
The study included 38 women during period 1 and 124 during period 2. The influenza diagnosis was confirmed for 24 of 38 (63.2%) women during period 1 and 65 of 124 (52.4%) women during period 2 (P = 0.24). The hospitalization rate fell significantly from period 1 to period 2, both in the total population (71.0% versus 44.3%, P = 0.004) and among women with confirmed influenza (83.3% versus 38.5%, P<0.001), as did the antibiotic therapy rate in both groups (respectively, 86.8% versus 56.1%, P = 0.001 and 91.7% versus 44.7%, P<0.001). The use of a rapid molecular assay for the diagnosis of influenza improved the management of pregnant women with an isolated fever or ILI by reducing the rates of unnecessary hospitalization and antibiotic therapy.
Fever during pregnancy is a frequent reason for an emergency consultation during the winter. After excluding obvious sites of infection, the most important diagnoses of isolated fever or influenza-like illness (ILI) to rule out are listeriosis and influenza—two infections that can lead to maternal or fetal complications and must be managed differently for pregnant women. The management of listeriosis during pregnancy is based on inpatient intravenous antibiotic therapy, while influenza in pregnancy, in the absence of signs of severity, is treated on an outpatient basis by oral antiviral drugs. The clinical signs of listeriosis are unspecific and sometimes resemble those of an ILI. Because clinical examination cannot distinguish listeriosis from influenza, the differential diagnosis depends on laboratory tests. While awaiting this laboratory confirmation, pregnant women with a fever or ILI are generally hospitalized for antibiotic therapy, because of the severity of the perinatal complications of listeriosis. These include late miscarriage or preterm delivery in more than 50% of cases and can lead to fetal or neonatal death in 25 to 50%. The bacteriological diagnosis of listeriosis is based on the identification of the bacteria from blood cultures, which take 48 to 72 hours to produce a result. Until recently, the molecular diagnosis of influenza required the separate technical steps of nucleic acid extraction and then amplification, most often performed in series only on weekdays during working hours, which delayed results. Automated individual rapid molecular assays of respiratory samples now provide results in less than 2 hours, which enables outpatient antiviral treatment and thus avoids unnecessary hospitalization and antibiotic therapy. Moreover, several studies have shown that the rapid start of antiviral treatment in pregnant women with influenza reduces the risk of maternal and fetal complications from this disease.[2,3]
This retrospective comparative before-and-after single-center study took place at the Port-Royal maternity hospital. The study population comprised the pregnant women with isolated fever or ILI who had a PCR test at the emergency department (ED) to diagnose influenza during the annual influenza epidemic periods from December 2012 through March 2017. These women were identified from the virology laboratory’s computerized register and crossed with the list of women hospitalized during the study periods for an infection to ensure that the women hospitalized for an unexplained fever had indeed been tested for influenza and listeriosis. Fever was defined as a temperature of 38°C or higher. ILI was defined by one or more of the following symptoms: a fever at home, shivering, muscle soreness or pain, headache, cough, or asthenia. All pregnant women presenting at the ED for fever had complete clinical and obstetric examinations and laboratory tests if needed. Women were excluded when the ED visit revealed that the fever was due to another evident infection site, such as a urinary tract infection or chorioamnionitis.
Overall, 162 women consulted for isolated fever after exclusion of other sites infection or ILI and had a diagnostic test for influenza during the study period: 38 during period 1 and 124 during period 2.
The use of a rapid molecular assay for influenza was associated with a reduction in the hospitalization rate for pregnant women consulting at the ED for isolated fever or ILI. This test thus makes it possible to avoid unnecessary hospitalization and antibiotic treatment in pregnant women with confirmed influenza, who account for more than half of this population. Moreover, determining the correct diagnosis—listeriosis or influenza—allows management appropriate for the actual maternal and fetal risks.
The use of a rapid molecular assay for the diagnosis of influenza during epidemics improved the management of pregnant women with an isolated fever or ILI by reducing the rates of unnecessary hospitalization and antibiotic therapy. These results confirmed our decision to adopt this rapid test in everyday practice.