2.1. Study design, period, and setting

The present multi-center cross-sectional study was conducted between March 21, 2020, and March 21, 2021, involving all pregnant women with COVID-19 who gave birth in public and private hospitals across three counties (Mahabad, Miandoab, and Bukan) located in the southern region of West Azerbaijan Province in northwest Iran. In this study, COVID-19 infection was confirmed in all pregnant women through positive RT-PCR test results at any stage of pregnancy, regardless of the presence or absence of clinical symptoms. The data were received and accessed for research purposes on March 22, 2021. All data were provided without any information that could identify individual participants, both during the data collection process and after its completion.

2.2. SARS-CoV-2 variant context

The study period (March 21, 2020, to March 21, 2021) corresponds to the initial waves of the COVID-19 pandemic in Iran, characterized by the prevalence of the original wild-type virus and the early Alpha (B.1.1.7) variant. Genomic surveillance studies have confirmed that the Delta (B.1.617.2) variant, which is linked to more severe outcomes, was first identified in the country in late 2021 [19,20].

2.3. The inclusion and exclusion criteria

The inclusion criteria for participants were:

1. 1). A confirmed positive RT-PCR test for COVID-19 during any stage of pregnancy.
2. 2). Having delivered by the time of the study.
3. 3). No prior receipt of the COVID-19 vaccine.

Participants were excluded if they met either of the following criteria:

1. Had not given birth by the time of the study (n = 44).
2. Experienced a miscarriage (n = 19).
3. Had incomplete medical or health records (n = 0).
4. Had baseline diseases (n = 108).
5. Were on regular medication (n = 50).
6. Had an induced PTB (n = 10)

The flowchart illustrating the sample selection process is shown in Fig 1.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Flowchart illustrating participant inclusion.

https://doi.org/10.1371/journal.pone.0341647.g001

2.4. Data collection

From the initial prenatal visit to the time of delivery, comprehensive data on baseline characteristics, clinical findings, as well as maternal and fetal outcomes, were systematically recorded in the National Health System. In this study, ‘hospitalization’ was defined specifically as admission required for the clinical management of COVID-19 symptoms (e.g., supplemental oxygen, management of severe respiratory symptoms, etc.). Hospital stays that were solely for obstetric indications (e.g., routine delivery, induction of labor) in women with COVID-19 were not counted as COVID-19-related hospitalizations in this study. This specific definition was applied to isolate the hospitalization burden attributable directly to SARS-CoV-2 infection and to avoid the potential confounding of hospitalization duration by delivery-related stays.

The following variables were obtained from electronic medical records in the National Health System: age (years), residency (rural, urban), education (illiterate, < diploma, diploma, > diploma), occupation (housekeeper, employed), symptoms (fever, cough, myalgia, chills, dysgeusia, fatigue, sore throat, headache, dyspnea, anosmia, chest pain, nausea, vomiting), parity (nullipara, multipara), and pre-pregnancy BMI (kg/m²).

2.5. Outcome variable

Maternal outcomes included hospitalization, PTB, mode of delivery, and stillbirth. Neonatal outcomes encompassed low birth weight (LBW), APGAR scores at 1 and 5 minutes, COVID-19 positivity, and neonatal death. PTB was defined as babies born alive before completing 37 weeks of pregnancy [21,22]. LBW refers to a newborn with a birth weight under 2500 g [23]. A stillbirth is characterized as the death of a fetus after 20 weeks of pregnancy but before or during delivery [24]. Also, neonatal death was defined as the death of a newborn within the first 28 days of life [25].

2.6. Statistical analyses

Categorical variables were presented as frequencies and percentages (N%). Continuous variables were summarized as mean ± standard deviation (SD) or median and interquartile range (IQR), based on their distribution. The normality of continuous data was verified using the Kolmogorov-Smirnov test. The association between categorical variables was assessed using the Chi-square test or Fisher’s exact test. Differences in continuous variables across the three groups were analyzed using the One-way analysis of variance (ANOVA) test for normally distributed data, with post-hoc comparisons (e.g., Bonferroni) conducted when significant differences were found; the Kruskal-Wallis test was used for non-normally distributed data. A p-value less than 0.05 was considered indicative of statistical significance. All statistical analyses were conducted using IBM SPSS Statistics for Windows, version 23.0 (Armonk, NY: IBM Corp).

2.7. Ethical consideration

This study was approved by the Research Ethical Review of Urmia University of Medical Sciences (approval number: IR.UMSU.REC.1399.352). The ethics committee of Urmia University of Medical Sciences waived the need for informed consent due to the retrospective nature of the study. All methods have been carried out in accordance with the relevant guidelines and regulations.

3.1. Demographic characteristics of pregnant women by trimester of COVID-19 infection

During the study, 217 pregnant patients with COVID-19 were included in the analysis: 20 (9.2%) were infected in the first trimester, 87 (40.1%) in the second trimester, and 110 (50.7%) in the third trimester. The overall mean age of participants was 29.54 ± 6.32 years. When analyzed by trimester, the mean age was 32.10 ± 6.06 years in the first trimester, 29.29 ± 6.32 years in the second trimester, and 29.28 ± 6.33 years in the third trimester. The majority of participants were housekeepers (93.1%) and resided in urban areas (84.3%). Regarding educational attainment, approximately half of the participants (51.2%) had an education level below diploma, while 23.0% had a diploma and 23.0% had university education. No statistically significant differences were found in demographic characteristics across trimesters (Table 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Demographic Characteristics of Participants by Pregnancy Trimester.

https://doi.org/10.1371/journal.pone.0341647.t001

3.2. Clinical characteristics of pregnant women with COVID-19

The clinical presentation of COVID-19 symptoms among pregnant women by infection trimester is detailed in Table 2. Analysis of 217 cases revealed that cough (52.1%) and myalgia (40.6%) were the most prevalent symptoms overall. A statistically significant difference was observed in the prevalence of fever across trimesters (p = 0.013), with increasing frequency from first trimester (10.0%) to third trimester (33.6%). No other symptoms showed significant variations across trimesters. The mean BMI across all participants was 27.04 ± 4.68 kg/m², falling within the overweight category according to WHO classification, with no statistically significant differences observed among trimesters (p = 0.586). The majority of symptoms including sore throat, dyspnea, chest pain, fatigue, chill, headache, nausea/vomiting, anosmia, and dysgeusia demonstrated comparable distribution patterns regardless of the trimester of infection.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Clinical characteristics and BMI of pregnant women by trimester of COVID-19 infection (N = 217).

https://doi.org/10.1371/journal.pone.0341647.t002

3.3. Maternal outcomes according to the trimester of COVID-19 infection

Maternal characteristics and obstetric outcomes were analyzed according to the trimester of COVID-19 infection among the 217 pregnant women included in the study. The overall mean gestational age at delivery was 38.13 ± 2.43 weeks, with no statistically significant difference observed across trimesters (p = 0.648). A significant association was found between the trimester of infection and PTB rates (p = 0.028), with the highest incidence observed following third-trimester infection (26.4%) compared to first (15.0%) or second (11.5%) trimester infections. The median duration of hospitalization was 0 days for all trimesters, with no significant differences found using either the Kruskal-Wallis test (p = 0.387) or the Median test (p = 0.712). No significant associations were detected between the trimester of infection and mode of delivery, or stillbirth rates. There were no maternal deaths recorded in any of the groups. These outcomes are summarized in Table 3.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Maternal characteristics and obstetric outcomes according to the trimester of COVID-19 infection (N = 217).

https://doi.org/10.1371/journal.pone.0341647.t003

3.4. Neonatal outcomes by infection trimester

Neonatal outcomes by infection trimester are summarized in Table 4. The mean birth weight of neonates born to women infected during the first, second, and third trimesters was 3185.00 ± 650.73, 3045.98 ± 561.31, and 2963.64 ± 529.80 grams, respectively (p = 0.215). The mean APGAR score at minute one for neonates showed statistically significant differences across trimesters, with scores of 9.08 ± 0.88, 8.45 ± 0.71, and 8.16 ± 0.78 for first, second, and third trimesters, respectively (p < 0.001). Similarly, APGAR scores at minute five also demonstrated significant differences, with means of 9.03 ± 0.90, 8.79 ± 0.83, and 8.39 ± 0.80 across the three trimesters (p < 0.001).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Neonatal outcomes by infection trimester (N = 217).

https://doi.org/10.1371/journal.pone.0341647.t004

Table 5 showed that post-hoc analysis with Bonferroni correction revealed a clear temporal gradient in Apgar scores. For Apgar1, pairwise comparisons showed significantly higher scores in first versus second trimester (mean difference = 0.621, 95% CI [0.167, 1.075], p = 0.003), first versus third trimester (mean difference = 0.921, 95% CI [0.475, 1.366], p < 0.001), and second versus third trimester (mean difference = 0.300, 95% CI [0.037, 0.562], p = 0.019). Similarly, for Apgar5, significant differences emerged between first and third trimesters (mean difference = 0.634, 95% CI [0.153, 1.115], p = 0.005) and between second and third trimesters (mean difference = 0.396, 95% CI [0.113, 0.680], p = 0.003), though no significant difference was observed between first and second trimesters (p = 0.731).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 5. Bonferroni post-hoc analysis of significant pairwise comparisons.

https://doi.org/10.1371/journal.pone.0341647.t005

4.1. Main findings

This multi-center cross-sectional study, conducted among a unique cohort of unvaccinated Iranian women during the initial wave of the pandemic, provides critical evidence on the trimester-specific impact of SARS-CoV-2 infection. This study reveals three principal findings. First, and most significantly, the timing of infection is a crucial determinant of PTB, with a pronounced risk gradient: women infected in the third trimester experienced a PTB rate of 26.4%, which is substantially higher than the 11.5% and 15.0% observed in the second and first trimesters, respectively. Second, while no vertical transmission was detected, neonatal condition at birth, as measured by Apgar scores at 1 and 5 minutes, was significantly lower in infants born to mothers with third-trimester infection compared to those infected earlier in pregnancy. Third, the clinical presentation and severity of COVID-19 varied with gestational age, evidenced by a higher prevalence of fever and a trend towards longer hospitalization durations in women infected during the third trimester. These findings underscore that late-pregnancy COVID-19 infection poses the greatest threat to both the course of pregnancy and immediate neonatal well-being in unvaccinated populations.

4.2. Comparison with the literature

Our core finding that third-trimester COVID-19 infection significantly elevates the risk of PTB resonates strongly with the global literature, thereby reinforcing the external validity of this association. A large population-based cohort study from Israel by Fallach et al. (2022) provided compelling evidence for this temporal relationship; infection during the third trimester, particularly after 34 weeks of gestation, was a strong independent risk factor for spontaneous preterm delivery [26]. Similarly, a comprehensive study from the United States by Stock et al. (2022) reported that pregnancies complicated by SARS-CoV-2 infection, especially during the Delta variant period, were associated with increased rates of stillbirth and PTB, with the latter being more common when infection occurred closer to term [7]. The pathophysiological rationale for this is well-grounded in the physiological adaptations of pregnancy. The third trimester is characterized by a state of relative immunodysregulation, decreased functional residual lung capacity, and a heightened pro-thrombotic state [10] SARS-CoV-2 infection exacerbates these conditions, potentially triggering a systemic inflammatory response that can lead to placental dysfunction and subsequently, uterine irritability or medical indications for early delivery [10].

The strength of our study lies in its demonstration of this risk within a distinct demographic context. The overall PTB incidence of 20.3% in our cohort is markedly higher than the pre-pandemic baseline rates of 9.2% to 12.4% reported for Iran in previous epidemiological studies [17,22]. This stark disparity suggests that SARS-CoV-2 infection conferred a substantial added burden on pregnancy outcomes in our population. Our findings are consistent with the international meta-analyses and cohort studies that established COVID-19 as a significant risk factor for PTB [7]. However, the magnitude of the effect we observed in the third trimester appears high. This may be attributed to the unique characteristics of our study population: all participants were unvaccinated.The absence of vaccine-induced immunity likely resulted in more severe maternal infection, which is a known driver of iatrogenic PTB [27].

Regarding neonatal outcomes, our finding of significantly lower Apgar scores at 1 and 5 minutes following third-trimester infection is a nuanced and important contribution. While many studies, including a meta-analysis by Neef et al (2021), have reported that most neonates born to infected mothers are asymptomatic and have good outcomes [13], few have performed a detailed trimester-based analysis of Apgar scores [14]. Our results suggest that even in the absence of confirmed vertical transmission or severe neonatal disease, the intrauterine environment of a mother with active late-pregnancy COVID-19 may be suboptimal, potentially due to subclinical placental insufficiency or transient maternal hypoxia affecting fetal oxygenation. This is supported by studies that have identified placental damage and inflammation in COVID-19 positive pregnancies, which can compromise fetal reserve and manifest as lower Apgar scores without progressing to stillbirth or neonatal death [11]. It is critical to interpret these findings in context; the mean 5-minute Apgar scores, while statistically lower, remained above 8 across all groups, indicating generally good neonatal condition. Nevertheless, the statistically significant decline warrants attention as a marker of fetal stress. Additionally, none of the newborns tested positive for COVID-19 immediately after birth. Similarly, a meta-analysis revealed that most neonates born to mothers with COVID-19 did not show any signs of positive infection [13].

Our analysis of maternal symptoms and hospitalization aligns with the existing understanding of COVID-19 severity in pregnancy. The trend of increasing hospitalization duration from the first to the third trimester, though not statistically significant in median values, reflects a pattern observed by Schell et al. (2022) and Sief et al (2023), who reported that disease progression and severity were more pronounced with later diagnosis [2,27]. The significant increase in fever prevalence during the third trimester in our cohort is a clinically relevant indicator of a more robust systemic inflammatory response, which could be a key mediator linking late-pregnancy infection to adverse outcomes like PTB [10]. The higher prevalence of nausea and vomiting in the first trimester is most parsimoniously explained by physiological pregnancy changes rather than a specific effect of SARS-CoV-2. Furthermore, the prevalence of gastroesophageal reflux disease (GERD) increases significantly as pregnancy progresses [28], which could confound the attribution of symptoms like nausea and vomiting to COVID-19 in later trimesters. While our data did not specifically differentiate GERD from other causes of nausea/vomiting, this physiological progression reinforces the importance of interpreting symptom data within the context of gestational age.

4.3. Translational implications for public health

The findings from this study carry significant translational implications for public health policy, clinical practice, and health systems planning, particularly in low- and middle-income countries (LMICs) with similar unvaccinated or under-vaccinated pregnant populations.

• Enhanced antenatal surveillance and triage: Our results argue for a trimester-stratified approach to managing COVID-19 in pregnancy. Public health guidelines should emphasize that pregnant women, especially those in their third trimester, constitute a high-risk group for severe COVID-19 and its complications. Symptomatic women presenting in late pregnancy should be prioritized for clinical assessment, monitoring for signs of preterm labor, and may require a lower threshold for hospitalization to manage potential respiratory compromise or disease progression.
• Vaccination advocacy and prioritization: Perhaps the most powerful public health implication is the reinforced critical importance of COVID-19 vaccination for pregnant women. Our data, derived from a pre-vaccination cohort, illustrate the “natural history” of the disease and its significant risks. These findings can be leveraged in public health campaigns to counteract vaccine hesitancy by providing clear, local evidence of the dangers of COVID-19 in pregnancy. Pregnant women should be actively prioritized in vaccination rollouts, with messaging that emphasizes protection against PTB and severe maternal morbidity, particularly during the high-risk third trimester.
• Resource allocation and health system preparedness: The increased rates of PTB and longer hospital stays associated with third-trimester infections have direct implications for resource allocation. Hospitals serving pregnant populations must ensure adequate capacity in obstetric wards, neonatal intensive care units (NICUs), and isolation facilities to manage potential surges in preterm deliveries and infected mothers. Prenatal care programs should integrate COVID-19 prevention education, early testing, and clear referral pathways for infected pregnant women.
• Long-term follow-up and research: The finding of lower Apgar scores, while not indicating severe immediate morbidity, raises questions about potential long-term neurodevelopmental outcomes. Public health systems should consider establishing registries for long-term follow-up of infants born to mothers with COVID-19. Further research is needed to investigate the mechanisms of placental damage and fetal impact, which could inform future therapeutic interventions.
• Implications for the inclusion of pregnant women in clinical research: The quantifiable and trimester-specific risks demonstrated in our findings, namely, the significantly elevated incidence of spontaneous PTB and the graded decline in neonatal Apgar scores associated with third-trimester SARS-CoV-2 infection highlight a critical imperative for the systematic inclusion of pregnant individuals in clinical trials for vaccines and therapeutics. The historical paradigm of excluding this population from research, ostensibly for fetal protection, has paradoxically engendered a critical evidence deficit, rendering them therapeutic orphans when interventions are deployed at scale [29–31]. Our data, which precisely delineate the severe maternal and neonatal morbidity in an immunologically naive cohort, provide empirical weight to the ethical argument against this exclusion. A rigorous risk-benefit analysis must now explicitly factor in the significant baseline risks of the disease itself, which our study helps to define. Consequently, the principle of justice in research necessitates the proactive enrollment of pregnant persons to generate safety, immunogenicity, and efficacy data relevant to their distinct pathophysiology [7,30]. Integrating pregnant women into clinical trial frameworks from the outset is not merely a logistical consideration but a moral and scientific necessity to ensure they are protected by evidence-based interventions in future public health emergencies, rather than remaining a vulnerable and understudied population.