Interpretation
Total number of COVID-19 cases diagnosed by the Italian Regional
Reference Laboratories among women of reproductive age was 28.661 in
Italy until April 28 with about 51% observed in the age group 40-49
years (10). In our study, 84.6% of cases were registered in the
northern regions, in line with the distribution on the total number of
COVID-19 cases in Italy, of whom 80% were observed in the North during
the study period (10).
Unlike studies regarding Chinese cases (11), which are retrospective and
hospital-based, this research, like the UKOSS study (4), is prospective
and population-based providing data to hypothesise several factors
associated with the development of the disease during pregnancy.
Medical history of the enrolled women shows that multiparas and obese
women are more frequent among those with COVID-19 infection compared to
the background population of women in reproductive age (12,13). The
prevalence of obesity is 17% vs 7% among Italian women of reproductive
age (14) and multiparas are 68% vs 49% of the women giving birth in
the northern Italian regions (15,16). Pregnant women with non-Italian
citizenship have the same prevalence reported in Northern Italy among
Italian residents in reproductive age (14,15); however, the proportion
presenting with pneumonia is higher compared to Italian women. This
difference could be due to a delayed access to care among immigrant
women. A higher than expected number of multiparas was affected,
probably because living with several children — who are often
asymptomatic — may facilitate transmission.
The most common symptoms reported at the time of hospitalisation were
fever (63%), cough (71%) and general weakness (48%) like those
described by the European Surveillance System (16) among the general
population and those reported by pregnant women affected by COVID-19 in
other countries (4,5,11).
COVID-19 interstitial pneumonia affected 41.5% of hospitalised women.
Previous comorbidities are significantly associated (p= 0.002) to the
risk of developing pneumonia. Comorbidities can drastically weaken the
immune system and can cause conditions directly related to a greater
risk of respiratory infections. Despite the small numbers of this
preliminary analysis, the 4 women with autoimmune disease deserve
attention.
The clinical picture of patients diagnosed with COVID-19 pneumonia is
similar to those described in China and UK (4,5,11), and it seems to be
less serious than the effects on the general population (16) although
this comparison should consider that the proportion of affected women
who access care is greater among pregnant women than the general
population. Similarly to the UKOSS study (4), among women affected by
COVID-19 pneumonia, 11% had severe complications and were admitted to
an ICU whereas among hospitalised patients with SARS-CoV-2, collected by
the Italian Integrated surveillance, patients requiring ICU admissions
were 22.8% (11). The lower number of women who developed severe
symptoms compared to the general population is in line with the
hypothesis that changes in the hormonal milieu in pregnancy, which
influence immunological responses to viral pathogens, together with the
physiological transition to a Type 2 T helper cells environment, which
favours the expression of anti-inflammatory cytokines, help to reduce
the inflammatory response that, in the very severe COVID-19 infections,
is responsible for multi-organ damage (17,18).
Compared to China (19), Italy uses X-rays more often than CT scans in
diagnosis, while lung sonograms are used both for pneumonia diagnosis
and monitoring. Validation of this diagnostic tool is of primary
importance because it is simple, cheap, safe, and easy for OBGYN staff
to access (20).
Lymphocyte counts and CRP are confirmed as the best laboratory
parameters for testing for the presence of the disease both in women
with pneumonia and in those without.
Effectiveness and safety of the drugs used are still uncertain due to
the small number of cases and to the limited available knowledge (21).
In the present cohort 76.7% of the hospitalised women received at least
one pharmacological treatment and 33.8% has been treated with an
antiviral (Tab.3). Treatments have been used in percentages considerably
higher among women affected by pneumonia, mainly hydroxychloroquine and
antivirals combined in approximately half of the cases alongside an
empirical antibiotic treatment. Hydroxychloroquine is the most widely
used drug, probably due to its ascertained safety profile in pregnant
women (21). Information on the use of anticoagulants is lacking and a
specific question has been recently added to the data collection form.
There are several open questions on clinical and public health policies
aspects that emerge from the data analysis and that merit further
research.
- Why do pregnant women affected by COVID-19 seem to be less at risk of
serious respiratory infections and unfavourable maternal outcomes than
in cases observed during the H1N1, MERS and SARS epidemics?
- What is the predictive value of various diagnostic methods using
images in diagnosing and monitoring interstitial pneumonia?
- How effective and safe are the drugs in use and what are their
clinical applications?
With a view to public health policy, it is essential to prepare for an
effective vaccination campaign against seasonal flu so that next fall
the uptake of flu jabs among pregnant women can be increased from
current levels of 7%. A further advantage of this would be to
distinguish more easily between seasonal flu and COVID-19. The
observation that almost half of the cohort had unaware at-risk contacts
in the two weeks before the onset of symptoms highlights the issue of
the asymptomatic infections impact and strengthens the recommendations
of social distancing and contact tracing measures amongst pregnant
women. Given the risk of asymptomatic viral shedding some hospitals
started universal screening of patients on labour and delivery. Defining
a single, nationwide strategy offering testing for pregnant women is of
paramount importance in order to avoid distortions in the estimates of
the frequency of the condition resulting from different regional
approaches.