Discussion
Main findings
The main finding of our study is a significantly higher rate of gestational diabetes in a SARS-CoV-2 infected pregnant population, when compared to historical controls, which raises concerns about the association between GDM and SARS-CoV-2 infection during pregnancy. All though no statistical correlation was found between the time point of infection in regards to OGTT, previous data on DM and COVID-19 during pregnancy would support in a first line that those patients with GDM are more prone to SARS-CoV-2 infection. On the other hand, multivariate regression analysis found BMI and COVID-19 to be independent risk factors for GDM in our cohort, supporting the theory of the virus-triggered diabetes onset. This is to our knowledge the first case-control study providing evidence, even if limited, for a possible causal relationship between COVID-19 and onset of GDM.
As stated before, the hyperglycemic level directly correlates with adverse obstetrical outcome [9,10]. What the severity of infection is concerned, to date, only a clear association between previously existing diabetes and severe course of SARS-CoV-2 infection could be established [18, 19]. Being able to determine the severity of the SARS-CoV-2 infection based on the NIH classification is one of the strengths of our study. All though the size of our cohort does not allow us to make a statement regarding the severity of SARS-COV-2 in relation to GDM, 50% of the women requiring ICU admission in our cohort suffered from GDM, which is alarming. On a deeper analysis, body mass index, GDM and time point of infection none correlated with inpatient management of SARS-CoV-2 infection, thus with the degree of severity (Table 4). Since previous large reports could clearly show a correlation between high BMI and severity of infection, we believe that our results are a consequence of the small number of women with inpatient management and ICU admission, thus lack of statistical power to demonstrate a possible association [19].
With an European rate of GDM of 16.3% and worldwide of up to 25.5%, these results are of concern and call for consequences in the management of pregnant patients suffering from GDM or at risk for GDM in the context of the pandemic [16].
A recently published multicentric study with similar design reports an association between insulin dependent GDM and COVID-19 diagnosis in pregnancy, yet over 80% of the participants were SARS-CoV-2 positive at the time-point of delivery, making the assessment of a bi-directional association difficult [14]. Until larger case-control studies are available, it remains open if this association is uni- or bidirectional, meaning if SARS-CoV-2 also plays a pathophysiological role in the genesis of GDM. Possible mechanisms of SARS-CoV-2 induced metabolic decompensation with onset of diabetes were described in the introduction of this report and are, at least on a theoretical basis, conceivable [9,10]. Several epidemiological studies note a seasonal distribution of gestational diabetes, which reinforces the ‘viral theory‘ of onset of diabetes, with extension to pregnant women [9,10].
In order to perform data analysis in our study, we matched the case population with historical controls based on parity, BMI and ethnicity. Since a high BMI and specific ethnicities are known risk factors for gestational diabetes, the rationale for choice of controls was to eliminate these cofounding factors from the analysis [10]. Furthermore, we decided for parity as a matching criterion with the intention to analyze presence of GDM in a previous pregnancy as a risk factor for the current GDM diagnosis (Table 2).
The rationale of choosing historical controls, i.e. pregnancies managed at our institution prior to the pandemic, was to secure that asymptomatic, not tested SARS-CoV-2 infected women were not included in the control group by accident. The timeline for control group was intentionally kept narrow (three years before the pandemic), in order to reduce bias that could possibly occur by fluctuations of GDM prevalence in time.
Although a further cofounding factor for SARS-CoV-2 infection in GDM affected women could be a higher exposition to hospital visits in these patients, we mention that management adaptation has been performed in our center during the major SARS-CoV-2 pandemic surges, i.e. reduction of consultations or conversion to telemedicine. Nevertheless, no alteration in diabetes testing regimens occurred, i.e. testing has been performed analogue to the pre-pandemic period. In both groups, women where OGTT was not available were excluded, in order to avoid possible diagnosis biasing. Homogeneity of testing is a major strength of our study, since standard OGTT was used in every single patient in both groups, which distinguishes us from previous publications.
In both our study groups, GDM rate was higher than in the general pregnant population in our country, which could be explained by the higher proportion of high-risk pregnancies as well as by the high number of South Asian immigrants being followed at our institution [16].
The rate of hospital admission in SARS-CoV-2 infection in our population was in line with previous reports [19]. We noted a significantly higher rate of premature delivery in the case group, as compared to the controls, where preterm delivery corresponds those of the general pregnant population [20]. The 17.33% rate of preterm delivery is in line with results from a large previous meta-analysis reporting 17% preterm delivery in SARS-CoV-2 infection during pregnancy [19].
Strengths and Limitations
One major strength of our study is the prospective data assessment in the case group and the case-control approach. As mentioned before, further strengths are represented by the homogeneity of GDM diagnosis in both groups, as well as the ability to classify the COVID-19 in respect to the symptoms. The major limitation is the cohort size as well as not having matched for further comorbidities or lower socioeconomic status because of incomplete records, which is a known risk factor for both GDM as well as SARS-CoV-2 infection [10,21].
Interpretation
According to CDC reports, only 31% of the pregnant US population has been vaccinated against SARS-CoV-2 so far [22]. We believe it is safe to extrapolate these numbers to the majority of the developed economy countries, although recommendations for vaccination in pregnant women have been issued in a considerable proportion of states worldwide [23]. Taking a deeper look at the vaccination rate of the general population in low-income countries, incidences of 12% for Asian countries or even significantly lower in the majority of African states have been reported [24]. All though pregnant women are generally considered to build a young and healthy population, increasingly high BMI and gestational diabetes rates, as well as maternal age, the wide spread of the highly contagious variants and still low vaccination acceptance make SARS-CoV-2 to a highly relevant issue in obstetrics.
Therefore, pregnant women in general and those with GDM in particular should not only be recommended early vaccination, but also caution in using protective measures. Moreover, appropriate counseling should be offered to these patients at risk.
Conclusions: The significantly higher rate of GDM among women with SARS-CoV-2 infection during pregnancy, as compared to matching controls, suggests that GDM increases the risk of infection. On the other hand, SARS-CoV-2 during pregnancy might increase the risk of developing GDM. Vaccination and caution in using protective measures should be recommended to pregnant women, particularly those with co-morbidities.
Disclosure statement: The author(s) report(s) no conflict of interest.