Confirmation of preeclampsia-like syndrome induced by severe COVID-19: an observational study

BACKGROUND Since the outbreak of the COVID-19 pandemic, some studies have reported an increased preeclampsia incidence in pregnant women with SARS-CoV-2 infection. Several explanations for this association have been proposed, including a preeclampsia-like syndrome induced by severe COVID-19. This syndrome was described in a small case series and has not been confirmed in larger studies, and its effect on perinatal outcomes has not been studied. OBJECTIVE This study aimed to confirm the preeclampsia-like syndrome because of COVID-19 and to investigate its implications on pregnancy outcomes and prognosis. STUDY DESIGN This was a prospective, observational study conducted in a tertiary referral hospital. The inclusion criteria were pregnant women admitted to the intensive care unit for severe pneumonia because of COVID-19. They were classified into 3 groups based on clinical and laboratory findings: preeclampsia, preeclampsia-like syndrome, and women without preeclampsia features. The 3 cohorts were analyzed and compared at 3 different times: before, during, and after severe pneumonia. The main outcomes were incidence of adverse perinatal outcomes and signs and symptoms of PE, such as hypertension, proteinuria, thrombocytopenia, elevated liver enzymes, and increased angiogenic factors (soluble fms-like tyrosine kinase 1–to–placental growth factor ratio). RESULTS A total of 106 women were admitted to the intensive care unit because of severe pneumonia, and 68 women were included in the study. Of those, 53 (50.0%) did not meet the diagnostic criteria for preeclampsia and remained pregnant after pneumonia (non-preeclampsia); 7 (6.6%) met the diagnostic criteria for preeclampsia, had abnormal (>38) soluble fms-like tyrosine kinase 1–to–placental growth factor ratio (preeclampsia), and delivered during severe pneumonia, and 8 (7.5%) met the diagnostic criteria for preeclampsia, had normal (≤38) soluble fms-like tyrosine kinase 1–to–placental growth factor ratio (preeclampsia like), and did not deliver during pneumonia. Despite not having delivered, most preeclampsia-related features improved after severe pneumonia in women with preeclampsia-like syndrome. Women with preeclampsia had significantly poorer outcomes than women with preeclampsia-like syndrome or without preeclampsia. CONCLUSION More than 50% of women with severe COVID-19 and diagnostic criteria for preeclampsia may not be preeclampsia but a preeclampsia-like syndrome, which may affect up to 7.5% of women with severe COVID-19. Preeclampsia-like syndrome might have similar perinatal outcomes to those of normotensive women with severe pneumonia because of COVID-19. For these reasons, preeclampsia-like syndrome should be excluded by using soluble fms-like tyrosine kinase 1–to–placental growth factor ratio in future research and before making clinical decisions.

T he etiology and pathogenesis of preeclampsia (PE) are not completely understood; however, deficient trophoblast invasion seems to play a leading role. 1 The resulting placental ischemia leads to an angiogenic imbalance in maternal serum, with increased antiangiogenic factors, such as soluble fms-like tyrosine kinase 1 (sFlt-1), and decreased proangiogenic factors, such as placental growth factor (PlGF). 2 Some studies have shown that this angiogenic status might induce endothelial cell dysfunction, which contributes to the development of PE with its signs and symptoms (hypertension, proteinuria, thrombocytopenia, elevated liver enzymes, and neurologic disturbances). 3 Several disorders have previously been demonstrated to mimic the clinical features of PE, including acute fatty liver, thrombotic thrombocytopenic purpura, hemolytic uremic syndrome, acute exacerbation of lupus erythematosus, and COVID-19. 4−7 These conditions share some clinical and laboratory findings with PE, and its diagnostic criteria may overlap; therefore, to provide adequate care, a differential diagnosis, although challenging, is necessary. Some studies have shown that the sFlt-1 −to−PlGF ratio is a highly specific marker for placental dysfunction and may help to accurately discriminate PE from some of its imitators. 6,8,9 Since the outbreak of the COVID-19 pandemic, some studies have reported an increased PE incidence in pregnant women with SARS-CoV-2 infection, and several explanations for this association have been proposed. 10−13 On May 2020, a small case series published by Mendoza et al 7 described a PE-like syndrome induced by severe COVID-19 during pregnancy. That study showed that up to 62.5% of pregnant women with severe pneumonia because of SARS-CoV-2 infection may go on to develop features of PE. Nevertheless, among the 5 cases with signs and symptoms of PE, only 1 case had abnormal angiogenic status, suggesting that the remaining 4 cases probably did not have PE. Of these 4 cases, only 1 patient was still pregnant after recovering from pneumonia, and all signs and symptoms of PE had disappeared. Given that PE does not resolve spontaneously, and delivery is the only definitive treatment, severe COVID-19-induced PE-like syndrome was the most plausible explanation for this case.
Therefore, only 1 confirmed case of PE-like syndrome was described in that study at the beginning of the COVID-19 pandemic; this syndrome has not been confirmed in larger studies, and its effect on perinatal outcomes has not been studied. Therefore, this study aimed to confirm the existence of a PElike syndrome in a larger cohort of women with severe COVID-19 and to investigate its implications for pregnancy outcomes and prognosis.

Materials and Methods
This was a prospective cohort study conducted at Vall d'Hebron Barcelona Hospital Campus, Barcelona, Spain, from February 2020 to September 2021. Our site was a referral hospital for severe COVID-19 pneumonia, especially among pregnant patients. The inclusion criteria were pregnant women admitted to the intensive care unit (ICU) for severe pneumonia because of SARS-CoV-2 infection during pregnancy. SARS-CoV-2 infection was confirmed by a real-time polymerase chain reaction of nasal and pharyngeal swabs. Participants who met the PE diagnostic criteria and had normal sFlt-1−to −PlGF ratio (≤38) were not delivered for PE and were classified as patients with PE-like syndrome. To establish whether the prognosis of women with PE-like syndrome was more similar to that of women with PE or to that of normotensive women, pregnancy outcomes were compared with 2 reference groups: women with true PE (met the PE criteria and had an sFlt-1−to−PlGF ratio of >38) and women who did not meet the PE criteria and did not deliver during the ICU stay (non-PE group). As comparing women with PE-like syndrome with women without PE and electively delivered because of pneumonia during the ICU stay did not provide additional information on the prognosis of patients with PE-like syndrome, these women were excluded from this study ( Figure).
Gestational age was confirmed by fetal crown-rump length measurement at the first-trimester scan. 14 Participants were examined at 3 different times: before severe pneumonia, during severe pneumonia (at the time of ICU stay), and after recovery from severe pneumonia. Demographic characteristics, obstetrical and maternal history, and biophysical and biochemical markers were collected. Biophysical and biochemical data included systolic blood pressure (SBP; mm Hg), diastolic blood pressure (DBP; mm Hg), hypertension (SBP of >140 mm Hg and/or DBP of >90 mm Hg), mean uterine artery pulsatility index (PI) percentile by transabdominal Doppler ultrasound, 15 proteinuria (in 24-hour urine samples, spot urine protein-to-creatinine ratio, and/or dipstick urinalysis), platelet count (mL), aspartate aminotransferase (AST; U/L), alanine aminotransferase (ALT; U/L), lactate

AJOG MFM at a Glance
Why was this study conducted?
The preeclampsia (PE)-like syndrome has not yet been confirmed in larger studies, and its effect on perinatal outcomes has not been studied.

Key findings
Despite not having delivered, most PE-related features in women with severe COVID-19 improved after severe pneumonia in women with PE-like syndrome. Women with PE had significantly poorer outcomes than women with PE-like syndrome or without PE. Angiogenic factors are useful to discriminate PE-like from true PE.
What does this add to what is known? PE-like syndrome may be accurately identified by angiogenic factors. Correct identification of cases with PE-like syndrome may improve pregnancy outcomes by reducing iatrogenic complications.
Original Research dehydrogenase (LDH; U/L), and creatinine level (mg/dL). In women with suspected PE, maternal serum PlGF and sFlt-1 (pg/mL) were determined using the fully automated Elecsys assays for sFlt-1 and PlGF on an electrochemiluminescence immunoassay platform (cobas e analyzers; Roche Diagnostics, Rotkreuz, Switzerland). The sFlt-1−to−PlGF ratio was calculated, with ≤38 values considered normal and unlikely to be associated with an underlying placental dysfunction. 16 PE was defined as new-onset high blood pressure (SBP of >140 mm Hg and/or DBP of >90 mm Hg), worsening of previous high blood pressure in addition to new-onset proteinuria (proteinto-creatinine ratio of >300), worsening of previous proteinuria, or at least 1 of the following signs and symptoms of severe PE: cerebral or visual symptoms, elevation of liver enzymes to twice the normal concentration, platelet count of <100.000/mL, serum creatinine level of >1.1 mg/dL, or pulmonary edema. 17 Here, to diagnose hypertension, at least 5 of 6 consecutive measurements had to be abnormal for a 6-hour period. Hemolysis, elevated liver enzymes, and low platelet count (HELLP syndrome) is considered a variant of PE. The diagnostic criteria for the HELLP syndrome were hemolysis with increased LDH (>600 U/L), AST or ALT (≥70 U/L), and platelet counts of <100 000/mL. 18 Elective delivery was recommended at ≥37 weeks of gestation in women with PE without severe features and ≥34 weeks of gestation in women with severe PE and/or HELLP syndrome. In COVID-19 cases that met the PE diagnostic criteria and had an sFlt-1−to −PlGF ratio of ≤38, elective delivery because of PE was not recommended; alternatively, these cases were managed according to the clinical evolution of COVID-19. Elective delivery because of severe pneumonia was indicated at ≥36 to 38 weeks of gestation and ≥33 to 35 weeks of gestation if invasive respiratory support was required.
This study was approved by the Vall d'Hebron Barcelona Hospital Campus Ethics Committee (PR[AMI]181/2020) on March 13, 2020. All patients included in the study gave their oral or written consent according to the criteria of our ethics committee.

Statistical analysis
Categorical data were reported as frequency and percentage, and comparisons among the groups were estimated by the chi-square or fisher tests, as appropriate. Continuous variables were reported as median (interquartile range [IQR]). The Mann-Whitney U test was used to assess differences among the groups. The Wilcoxon signed-rank test was used to assess differences of repeated measurements on a single sample. The statistical significance level was set at P<.05, and the Bonferroni correction was applied to compensate for multiple comparisons. The Stata statistical software package was used for data analysis (StataCorp 2019; Stata Statistical Software: Release 16; StataCorp LLC, College Station, TX).

Results
During the study period, 106 patients with severe pneumonia because of COVID-19 were admitted to the ICU. Of those, 38 (35.8%) did not meet the inclusion criteria for needing elective delivery during the ICU stay because of clinical worsening of pneumonia and not having clinical suspicion of PE. Therefore, 68 women (64.2%) were included in the study: 7 (6.6%) with PE, 8 (7.6%) with PE-like syndrome, and 53 (50.0%) without PE features and not needing delivery during severe pneumonia (non-PE group). All    .  .045

Original Research
We found that some PE-related features worsened similarly during severe pneumonia across all 3 groups; however, most of these features improved significantly after recovery from severe pneumonia, without requiring elective delivery in most cases (Tables 2 and 3). Among the 68 women included in this study, there were 15 cases (22.1%) of hypertension, 19 cases (27.9%) of proteinuria, and 43 cases (63.2%) of elevated liver enzymes. There was no case of thrombocytopenia or abnormal LDH or creatinine. Moreover, 15 women (22.1%) met the PE diagnostic criteria during severe pneumonia. When comparing the incidence of PErelated features among groups, hypertension was present in the PE and PElike syndrome groups (100%) and not in the non-PE group (0%). In addition, the sFlt-1−to−PlGF ratios were higher in the PE group than in the PE-like syndrome and non-PE groups ( When we compared the groups after the resolution of severe pneumonia because of COVID-19, only blood pressure was significantly different among groups, being higher in the PE group, although always within the normal range ( No difference was found between the PE-like syndrome and non-PE groups. More information can be found in Table 5.

Principal findings
During severe pneumonia, up to 14.2% (15/106) of women met the PE diagnostic criteria; however, 8 of 15 cases (53.3%) improved after pneumonia without elective delivery. These 8 cases had normal sFlt-1−to−PlGF ratios, better outcomes than the 7 women with PE, and similar pregnancy outcomes to those of normotensive women with severe COVID-19 who did not deliver during severe pneumonia. This suggests that the sFlt-1−to−PlGF ratio may be a useful marker to identify a subset of patients with PE-like syndrome that otherwise may be misdiagnosed with PE and who would benefit from a more conservative management.
In Table 2, we confirm that in both the PE-like syndrome group and the PE group, most PE features observed during severe pneumonia improved after recovery, especially hypertension and transaminitis. Improvement in the PE group can be attributed to the fact that all women delivered because of COVID-19 and/or PE during ICU stay. In contrast, in the PE-like group, PErelated features improved without delivery. Given that PE does not resolve spontaneously, but only after delivery, the most plausible explanation for this outcome in the PE-like group is a PElike syndrome induced by severe COVID-19.
In the non-PE group, some PErelated features (such as increased liver enzymes) were observed during severe pneumonia but improved after recovery. This suggests that even in normotensive patients, severe pneumonia because of COVID-19 can induce transient systemic effects that mimic PE and that resolve completely after the resolution of pneumonia.
During severe pneumonia (Table 3), the PE-like and PE groups were identical, except for sFlt-1−to−PlGF and creatinine values that were significantly lower, although within the normal range, in all patients with PE-like syndrome. However, creatinine values were also normal in the PE group, making this finding clinically irrelevant for differential diagnosis. Therefore, the sFlt-1 −to−PlGF ratio seems to be the only parameter that allows to accurately discriminate PE from PE-like syndrome.
All women with PE were electively delivered during severe pneumonia. After resolution of severe pneumonia (Table 4), most PE-related features, such as hypertension, transaminitis, and increased sFlt-1−to−PlGF ratio, improved similarly across all groups, with the remarkable difference that all cases in the PE-like syndrome and non-PE groups recovered without elective delivery. This confirms that PE-related features in these 2 groups were due to the systemic effects of COVID-19.
Regarding perinatal outcomes (Table 5), the rate of cesarean deliveries was significantly higher and the birthweight and gestational age at delivery were significantly lower in the PE group than in the PE-like and non-PE groups. No difference was found between the PE-like and non-PE groups for perinatal outcomes.

Results
Several studies have reported an association between PE and COVID-19 that may be explained by several etiopathogenic pathways. 13,19−22 PE-like syndrome was described at the beginning of the pandemic in a small case series and could partly explain the higher PE

Research implications
In addition, our results have important research implications, as we show that some cases classified as true PE in previous studies might instead be cases with PE-like syndrome. Therefore, to better assess the association between true PE and COVID-19, in future studies, PElike syndrome should be excluded using the sFlt-1−to−PlGF ratio.

Strengths and limitations
One of the main strengths of this study was the large number of pregnant women with severe pneumonia because of COVID-19. In addition, this study confirmed the existence of a PE-like syndrome induced by severe COVID-19 by assessing the sFlt-1−to−PlGF in a relatively large cohort of pregnant women with suspicion of PE and severe COVID-19. This study provided novel evidence for the evolution and outcomes of pregnancies with PE-like syndrome induced by severe COVID-19. Another strength of this study was that it further supports the clinical usefulness of the sFlt-1−to−PlGF ratio for discriminating between placental dysfunction and other disorders. 4 The main limitation of this study was the reduced number of cases with PE and PE-like syndrome, which makes it harder to identify differences across groups for pregnancy outcomes. For instance, the median birthweight in the PE-like syndrome group tended to be lower than in the non-PE group (2737.5 g vs 3130 g, respectively), despite gestational age at delivery being the same. However, this difference was not statistically significant (P=.09), probably because of the reduced sample size in the PE-like syndrome group. Therefore, it is plausible that patients who are hypertensive with severe COVID-19 and normal sFlt-1−to −PlGF ratio (PE-like group) may have poorer perinatal outcomes than normotensive patients with severe COVID-19 (non-PE group). However, this hypothesis could not be confirmed in this study. Another limitation was that mean uterine artery PI was not recorded in any women in the PE group; for this reason, we could not investigate its role in distinguishing PE from PE-like syndrome.

Conclusions
More than 50% of women with severe COVID-19 that meet the diagnostic criteria for PE have PE-like syndrome. Therefore, future research on COVID-19 and PE should identify the cases with PE-like syndrome to better assess the association of both conditions. PElike syndrome is a transient condition with similar pregnancy outcomes to those of normotensive women with severe COVID-19. Therefore, correct identification of cases with PE-like syndrome may improve pregnancy outcomes by reducing iatrogenic complications. &