In this article, several errors were identified due to two overlapping studies in a meta-analysis. The authors have re-done their analysis to remove the duplicated data and the corrected text is given below.

In the abstract methods, the text that read:

We searched the PubMed, EMBASE, and Cochrane CENTRAL databases up to 15 December 2023,

We searched the PubMed, EMBASE, and Cochrane CENTRAL databases up to 13 December 2023,

In the abstract methods, the text that read:

The literature search identified 18,334 potential records, and 22 studies involving 3,042,997 pregnant women were ultimately included. Compared with the unexposed group, quetiapine [odds ratio (OR), 1.19; 95% credible interval (CrI), 1.01–1.39], aripiprazole (OR, 1.30; 95% CrI 1.10–1.65), olanzapine (OR, 1.33; 95% CrI 1.11–1.64), risperidone (OR, 1.43; 95% CrI 1.18–1.77), and lithium (OR, 1.61; 95% CrI 1.07–2.30) were associated with a slightly increased risk of congenital malformations. In contrast, lamotrigine (OR, 1.21; 95% CrI 0.86–1.64), ziprasidone (OR, 1.14; 95% CrI 0.73–1.72), and haloperidol (OR, 1.26; 95% CrI 0.90–1.75) did not show significant differences compared with the unexposed group, with narrower credible intervals.

The literature search identified 18,334 potential records, and 21 studies involving 1,701,089 pregnant women were ultimately included. Compared with the unexposed group, aripiprazole [odds ratio (OR), 1.29; 95% credible interval (CrI), 1.001–1.93] was associated with a slightly increased risk of congenital malformations. In contrast, lamotrigine (OR, 1.19; 95% CrI, 0.72–1.85), quetiapine (OR, 1.13; 95% CrI, 0.81–1.44), olanzapine (OR, 1.33; 95% CrI, 0.98–1.87), risperidone (OR, 1.35; 95% CrI, 0.96–1.93), and haloperidol/penfluridol (OR, 1.43; 95% CrI, 0.90–2.17) did not show significant differences compared with the unexposed group, with narrower credible intervals.

In the abstract Conclusion, the text that read:

The evidence from this analysis suggests that, overall, quetiapine has the lowest teratogenic risk when used during pregnancy, making it the safer option for pregnant women. Lamotrigine and haloperidol follow closely behind.

The evidence from this analysis suggests that, overall, quetiapine is currently the most rational choice among second-generation antipsychotics regarding safety and clinical utility. Among mood stabilizers, lamotrigine appears to be a relatively safer option. Haloperidol remains a viable alternative under close monitoring.

In Section 2.6 of the Methods, the text that read:

Therefore, we used CINeMA, an improved online application based on GRADE principles, to assess the level of evidence of the 22 non-randomized studies included.

Therefore, we used CINeMA, an improved online application based on GRADE principles, to assess the level of evidence of the 21 non-randomized studies included.

In Section 3.1 of the Results, the text that read:

In total, the full texts of 354 studies were examined, and 332 studies were excluded. Twenty-two studies provided data on at least one outcome and were included in the quantitative synthesis.

In total, the full texts of 354 studies were examined, and 333 studies were excluded. Twenty-one studies provided data on at least one outcome and were included in the quantitative synthesis.

In Section 3.2 of the Results, the text that read:

We included 22 observational studies (3,042,997 patients) published from 1983 to 2023. The number of patients included in each study ranged from 58 to 1,341,544.

A total of 68.1% of the studies examined women in early pregnancy, while 27.2% of the studies examined women throughout pregnancy.

Fifty percent of the studies included in the Institute included controls who had experienced psychiatric disorders during pregnancy but had not taken antipsychotic medication. Sixteen studies (3,040,761 participants) reported data for the primary outcome (congenital malformations).

We included 21 observational studies (1,701,089 patients) published from 1983 to 2023. The number of patients included in each study ranged from 7 to 1,322,955.

A total of 66.7% of the studies examined women in early pregnancy, while 28.6% of the studies examined women throughout pregnancy.

A total of 47.6% of the studies included in the review included controls who had experienced psychiatric disorders during pregnancy but had not taken antipsychotic medication. Fifteen studies (1,698,670 participants) reported data for the primary outcome (congenital malformations).

In Section 3.3.1 of the Results, the text that read:

The NMA of overall congenital malformations included 16 cohort studies in which 12 interventions were assessed and 3,040,761 pregnant women were included.

Quetiapine, aripiprazole, and olanzapine were included in the largest number of comparison studies, followed by risperidone. However, lurasidone, valproate and carbamazepine were mentioned in only a small number of studies.

Compared to the unexposed group, the quetiapine (OR, 1.19; 95% CrI 1.01–1.39), aripiprazole (OR, 1.3; 95% CrI 1.1–1.65), olanzapine (OR, 1.33; 95% CrI 1.11–1.64)), risperidone (OR, 1.43; 95% CrI 1.18–1.77), and lithium (OR, 1.61; 95% CrI 1.07–2.3) groups all had a significantly increased risk of congenital malformations. In contrast, there was no difference between the other seven medication groups and the control group. Among these drugs, the drugs with narrower CrI were lamotrigine (OR, 1.21; 95% CrI 0.86–1.64), ziprasidone (OR, 1.14; 95% CrI 0.73,1.72) and haloperidol/penfluridol (OR, 1.26; 95% CrI 0.9, 1.75). These treatments were at least as effective as those in the unexposed group.

The NMA of overall congenital malformations included 15 cohort studies in which 12 interventions were assessed and 1,698,670 pregnant women were included.

Quetiapine and olanzapine were included in the largest number of comparison studies, followed by aripiprazole, risperidone and lithium. However, lurasidone, ziprasidone, valproate and carbamazepine were mentioned in only a small number of studies.

Compared with the unexposed group, the aripiprazole (OR, 1.29; 95% CrI 1.001–1.93) group had a significantly increased risk of congenital malformations. In contrast, there was no difference between the other eleven medication groups and the control group. The estimates for quetiapine (OR 1.13; 95% CrI 0.81–1.44), olanzapine (OR 1.33; 95% CrI 0.98–1.87), risperidone (OR 1.35; 95% CrI 0.96–1.93), lamotrigine (OR 1.19; 95% CrI 0.72–1.85), and haloperidol/penfluridol (OR 1.43; 95% CrI 0.90–2.17) were the most precise (i.e., had narrower CrIs), indicating more robust evidence for these comparisons. These treatments showed a safety profile comparable to the unexposed group regarding congenital malformations.

In Section 3.3.2 of the Results, the text that read:

In terms of congenital anomaly outcomes, the safest treatment, except for no exposure (SUCRA = 0.11), was lurasidone (SUCRA = 0.32), followed by ziprasidone (SUCRA = 0.33), quetiapine (SUCRA = 0.34), lamotrigine (SUCRA = 0.39), and haloperidol/penfluridol (SUCRA = 0.45). Carbamazepine had the highest risk (SUCRA = 0.87).

In terms of congenital anomaly outcomes, the safest treatment, except for no exposure (SUCRA = 0.15), was quetiapine (SUCRA = 0.28), followed by lamotrigine (SUCRA = 0.37). Conversely, carbamazepine had the highest risk (SUCRA = 0.85), followed by valproate (SUCRA = 0.77).

In Section 3.6 of the Results, the text that read:

We assessed 22 cohort studies using the ROBINS-I risk of bias assessment tool (see Tables 1 and 2 in Supplement 3). Overall, these studies exhibited a moderate to high risk of bias (specifically, 9 studies were classified as moderate risk and 13 studies as serious risk).

The quality of evidence for the primary outcomes, assessed using the CINeMA tool, ranged from moderate to very low (specifically, 6 comparisons were rated as moderate, 16 as low, and 58 as very low).

The comparisons classified as moderate risk involved control groups compared with quetiapine, aripiprazole, olanzapine, lithium, risperidone, and the comparison between aripiprazole and quetiapine.

We assessed 21 cohort studies using the ROBINS-I risk of bias assessment tool (see Tables 1 and 2 in Supplement 3). Overall, these studies exhibited a moderate to high risk of bias (specifically, 8 studies were classified as moderate risk and 13 studies as serious risk).

The quality of evidence for the primary outcomes, assessed using the CINeMA tool, ranged from moderate to very low (specifically, 7 comparisons were rated as moderate, 10 as low, and 61 as very low).

The comparisons classified as moderate risk involved control groups compared with aripiprazole, olanzapine, lithium, risperidone and haloperidol/penfluridol, the comparison between aripiprazole and quetiapine, as well as between quetiapine and lithium.

In Section 4 of the Discussion, the text that read:

We included a total of 22 studies covering 12 commonly used antipsychotics and mood stabilizers. The results indicated that for the primary outcome—the risk of congenital malformations—the treatment groups for quetiapine, risperidone, aripiprazole, olanzapine, and lithium had a higher risk of fetal congenital malformations compared to the unexposed group. In contrast, the following drugs did not show a statistically significant risk to fetal development compared to the control group: lurasidone (SUCRA = 0.32), ziprasidone (SUCRA = 0.33), lamotrigine (SUCRA = 0.39), and haloperidol (SUCRA = 0.45), and these drugs ranked relatively high in safety.

We included a total of 21 studies covering 12 commonly used antipsychotics and mood stabilizers. The results indicated that for the primary outcome—the risk of congenital malformations—the treatment groups for aripiprazole had a higher risk of fetal congenital malformations compared to the unexposed group. In contrast, the following drugs did not show a statistically significant risk to fetal development compared with the control group: quetiapine (SUCRA = 0.28) and lamotrigine (SUCRA = 0.37) and these drugs ranked relatively high in safety, whereas carbamazepine (SUCRA = 0.85) and valproate (SUCRA = 0.77) ranked lower in safety.

In Section 4.1 of the Discussion, the text that read:

Our network meta-analysis (NMA) showed that most second-generation antipsychotics, such as quetiapine, aripiprazole, olanzapine, and risperidone, slightly increased the risk of congenital malformations compared with the control group. The increase in risk was small for all drugs; even for risperidone, which had the highest risk, the odds ratio (OR) was less than 1.5 times higher than the control group. This is below the commonly cited threshold of a twofold increase in risk for clinical significance in this field [45]. Based on SUCRA values, quetiapine (SUCRA = 0.34) was significantly less risky than the other three drugs (with SUCRA values of 0.50, 0.54, and 0.65, respectively). Consistent with our findings, a 2023 cohort study that included 21,751 participants using second-generation antipsychotics found that among the four drugs studied, quetiapine (absolute risk: 4.24%) and aripiprazole (absolute risk: 4.16%) had lower risks [46]. However, because quetiapine had a narrower confidence interval, its risk estimate may be more precise, which could also be attributed to the larger sample size for quetiapine (n = 11,065) compared to aripiprazole (n = 4523).

Another cohort study (n = 365) indicated that quetiapine does not significantly increase the risk of congenital malformations [47]. Although this finding is not consistent with our results, the study also noted that due to the limited sample size, it could only rule out more than a fivefold increase in risk, similar to the known teratogenic risk of valproate. Therefore, a slight increase in risk may have gone undetected [47]. In our study, the total sample size for quetiapine was 18,545 individuals. The results showed that, compared with the control group, quetiapine had an odds ratio (OR) of 1.19 with a 95% credible interval (CrI) of 1.01–1.38, suggesting a slight increase in the risk of congenital malformations. However, since the OR for quetiapine is only slightly above 1 and the lower limit of the credible interval is close to 1 (1.01), this slight increase in risk may not be sufficient to change clinical practice, especially when the potential therapeutic benefits of quetiapine outweigh the risks.

Among the evaluated second-generation antipsychotics, quetiapine shows a relatively low teratogenic risk and is also the most extensively studied antipsychotic to date. This finding is clinically important and is highly consistent with the results of a large body of observational studies. For example, a large cohort study involving 11,065 infants exposed to quetiapine showed that quetiapine exposure did not significantly increase the relative risk of congenital malformations, and this pattern was consistent across specific malformation subtypes [45]. In addition, after analyzing 264 pregnancies, Cohen and colleagues similarly concluded that quetiapine lacks evident teratogenicity [46]. This favorable safety profile is important for clinical practice, particularly given quetiapine’s central role in treatment guidelines.

New text in Section 4.1 of the Discussion, should read:

However, in our analysis, aripiprazole (OR 1.29; 95% CrI 1.001–1.93) was associated with a statistically significant increase in the odds of congenital malformations compared with no exposure. This finding contrasts with Huybrechts et al. [45], who reported no significant association based on propensity-score adjusted analyses. Our estimates reflect pooled unadjusted data from the included studies. Therefore, this result should be interpreted in the context of the methodological differences between the unadjusted NMA estimates and adjusted cohort analyses.

In Section 4.1 of the Discussion, the text that read:

Due to their relatively short time in clinical use, data on these drugs are limited, with 137 and 729 infants exposed, respectively. Although these two drugs rank first and second in SUCRA rankings, this does not indicate that they are safe in terms of teratogenic risk [52].

Due to their relatively short time in clinical use, data on these drugs are limited, with 137 and 32 infants exposed, respectively. Although lurasidone ranks fourth in SUCRA rankings, this does not indicate that it is safe in terms of teratogenic risk [52].

In Section 4.2 of the Discussion, the text that read:

These results are consistent with our study, which shows that the odds ratio for congenital malformations associated with haloperidol use is 1.259 (95% CrI 0.894–1.746).

These results are consistent with our study, which shows that the odds ratio for congenital malformations associated with haloperidol use is 1.43 (95% CrI 0.90–2.17).

In Section 4.3 of the Discussion, the text that read:

The outcomes for the lamotrigine group were similar to those of the unexposed group, while lithium significantly increased the risk of congenital malformations. Due to insufficient sample sizes and low evidence quality, this study could not reliably determine the risks associated with carbamazepine and valproate.

The outcomes for the lamotrigine group were similar to those of the unexposed group. Due to insufficient sample sizes and low evidence quality, this study could not reliably determine the risks associated with lithium, carbamazepine and valproate.

In Section 4.5 of the Discussion, the text that read:

As a result, our study included only cohort studies, which may introduce greater bias and result in lower overall evidence quality.

As our analysis relied on aggregate data from cohort studies, the pooled estimates are unadjusted and subject to residual confounding. Therefore, results should be interpreted with caution, focusing on the comparative rankings of interventions rather than absolute effect sizes.

In Table 1, the study by Huybrechts et al. was removed due an overlapping data set [1]. Additionally, due to sample overlap between the Cohen and Freeman studies, the aripiprazole intervention arm was removed from the Cohen study data [2, 3]; however, the Cohen study itself remains in the table.

For completeness and transparency, both the old incorrect and the corrected versions of Table 1 are shown below.

Incorrect Table 1

Table 1 Description of characteristics of included studies

Author, Year

Design

Indication

Nonexposed group

Interventions

Sample

Source

Age, years (mean ± SD/range)

Time of exposure

Duration of follow-up

Outcomes

Study conclusion

Huybrechts KF (2023) [46]

Cohort study

Psychiatric disorders

Unexposed with psychiatric disorders

Quetiapine,

aripiprazole,

olanzapine,

risperidone,

haloperidol/

penfluridol

341,363

FGAs1: 28.54

SGAs2: 29.54

NEG: 27.54

First trimester

NR

Major congenital malformations.

(1) Overall, in utero exposure to antipsychotics was not significantly associated with an increased risk of malformations.

(2) Olanzapine may be associated with an increased risk of oral clefts, atypical antipsychotics with an increased risk of gastroschisis and brain anomalies, and chlorprothixene with an increased risk of cardiac malformations, but further research is needed.

Freeman MP (2021) [27]

Cohort study

Psychiatric disorders

Unexposed with psychiatric disorders

Aripiprazole

867

EG (ari):

32.4 ± 5.49

NEG: 32.7 ± 4.19

First trimester

The follow-up was conducted at 7 months of pregnancy and 12 weeks postpartum.

Major congenital malformations.

The risk of major malformations following early pregnancy exposure to aripiprazole was not significantly increased, but the findings are limited by the relatively small sample size. Larger studies are needed in the future to further validate these results.

Ellfolk M, 2021) [26]

Cohort study

Psychiatric disorders

Unexposed with psychiatric disorders

Quetiapine,

aripiprazole,

olanzapine,

risperidone,

clozapine

26,139

NR

First trimester

NR

Major congenital malformations.

Olanzapine use was associated with an increased risk of major congenital malformations and specifically, musculoskeletal malformations.

Cohen LS (2018) [47]

Cohort study

Psychiatric disorders

Unexposed with psychiatric disorders

Quetiapine

365

EG (que):

32.3 ± 4.89

NEG: 33.3 ± 4.07

First trimester

The follow-up was conducted at 7 months of pregnancy and 12 weeks postpartum.

Major congenital malformations.

Quetiapine was not a major teratogen. However, considering the uncertainty in risk estimation due to sample size limitations, the study results could only rule out approximately a fivefold increase in the risk of major malformations.

Bellet F (2015) [8]

Cohort study

Psychiatric disorders

Unexposed but

unknown disease status

Aripiprazole

232

EG (ari):

31.8 ± 5.8

NEG: 31.4 ± 5.4

4–10 gestational weeks

The follow-up was conducted within 2 months after the expected delivery date.

(1) All congenital malformations (both major and minor).

(2) Preterm birth.

(3) Miscarriage.

No significant association was found between embryonic exposure to aripiprazole and major malformations. However, larger prospective studies are needed to further elucidate the reproductive safety of aripiprazole.

Huybrechts KF (2016) [11]

Cohort study

Psychiatric disorders

Unexposed with psychiatric disorders

Quetiapine,

aripiprazole

olanzapine,

risperidone,

ziprasidone

1,341,544

SGAs2:

25.39 ± 6.42

NEG:

24.01 ± 5.77

First trimester

NR

All congenital malformations.

Early pregnancy use of antipsychotics (APs) generally did not significantly increase the risk of congenital malformations, particularly cardiac malformations. However, the slight increase in malformation risk observed with risperidone warranted further investigation.

McKenna K (2005) [69]

Cohort study

Psychiatric disorders

Not exposed to teratogenic drugs, no mental illness or use of psychotropic drugs

Olanzapine

185

NR

First trimester

The follow-up was conducted 3 to 4 months after the expected delivery date.

Major congenital malformations.

Atypical antipsychotics did not appear to be associated with an increased risk of major malformations.

Diav-Citrin O (2005) [61]

Cohort study

Psychiatric disorders

Not exposed to teratogenic drugs

Haloperidol/penfluridol

709

EG (hal/pen):

32 (28–36)

NEG: 30 (27–33)

First trimester

The follow-up could extend up to 6 years, but most follow-ups were completed within the first 2 years after the infant's birth.

(1) Major congenital malformations.

(2) Miscarriage.

(3) Preterm birth.

Pregnant women using haloperidol or penfluridol had a higher risk of preterm birth compared to the control group, but there were no significant differences in the risks of congenital malformations and miscarriage.

Patorno E (2017) [70]

Cohort study

Psychiatric disorders

Unexposed with psychiatric disorders

Lamotrigine,

lithium

1,325,563

EG (lit)3:

25.6 ± 6.1

NEG: 24.0 ± 5.8

First trimester

NR

All congenital malformations.

Lithium use by mothers in the first trimester of pregnancy was associated with an increased risk of cardiac malformations.

Cohen LS (2023) [55]

Cohort study

Psychiatric disorders

Unexposed with psychiatric disorders

Quetiapine,

lurasidone,

aripiprazole

1477

EG (lur):

33.2 ± 4.92

EG (que):

32.0 ± 4.80

NEG: 32.6 ± 4.18

First trimester

The follow-up was conducted at 7 months of pregnancy and 12 weeks postpartum.

Major congenital malformations.

Lurasidone and quetiapine did not appear to be major teratogens, but additional information is needed to refine the risk estimates.

Habermann (2013) [71]

Cohort study

Psychiatric disorders

NR

Aripiprazole, quetiapine,

olanzapine, risperidone,

clozapine, ziprasidone

447

SGAs2:

32 (27–36)

First trimester

The follow-up was conducted 8 weeks after the delivery date.

Major congenital malformations.

The study did not find that second-generation antipsychotics (SGAs) posed a significant teratogenic risk.

Kulkarni J (2014) [72]

Cohort study

Psychiatric disorders

NR

Clozapine,

olanzapine,

quetiapine,

risperidone,

124

SGAs2:

32.67 ± 4.7

First trimester

The follow-up was conducted every 6–8 weeks during pregnancy.

Congenital malformations.

The rate of congenital anomalies in Australia was higher than the expected rate, particularly for congenital heart defects.

Viguera, A.C (2023) [73]

Cohort study

Psychiatric disorders

Unexposed with psychiatric disorders

Olanzapine

1207

EG (ola):

33.6 ± 4.56

NEG: 32.5 ± 4.08

First trimester

The follow-up was conducted at 7 months of pregnancy and 12 weeks postpartum.

Major congenital malformations.

Exposure to olanzapine in early pregnancy did not significantly increase the risk of major malformations, the study found.

Bodén R (2012) [74]

Cohort study

Bipolar disorder

NR

Lamotrigine,

lithium,

valproate,

carbamazepine

266

NR

During pregnancy

NR

Major congenital malformations.

Infants born to women with bipolar disorder had an increased risk of preterm birth, regardless of whether the mother received mood stabilizer treatment. Infants of untreated women also faced a higher risk of microcephaly and neonatal hypoglycemia.

Diav-Citrin O (2014) [75]

Cohort study

Psychiatric disorders

Unexposed with psychiatric disorders

Lithium

184

EG (lit):

32 ± 6

NEG: 31 ± 6

During pregnancy

The follow-up was conducted within the first 2 years after birth in most cases.

(1) Major congenital malformations

(2) Miscarriage.

(3) Preterm delivery.

Lithium treatment during pregnancy was associated with a higher incidence of fetal cardiovascular abnormalities.

Källén B (1983) [76]

Cohort study

Psychiatric disorders

Exposed to psychotropic drug but unexposed to lithium with psychiatric disorders

Lithium

79

NR

First trimester

NR

Malformations.

The use of lithium in early pregnancy was found to possibly increase the risk of perinatal mortality and malformations in infants; therefore, it is recommended to avoid lithium during early pregnancy.

Vigod SN (2015) [9]

Cohort study

Psychiatric disorders

NR

Quetiapine,

olanzapine,

risperidone

834

SGAs2:

28.8 ± 6.1

NEG: 26.7 ± 6.3

During pregnancy

NR

Preterm birth.

Antipsychotic drug use in pregnancy had a minimal evident impact on important maternal medical and short-term perinatal outcomes. However, the rate of adverse outcomes is high enough to warrant a careful assessment of maternal and fetal well-being among women prescribed an antipsychotic drug in pregnancy.

Sørensen MJ (2015) [77]

Cohort study

Psychiatric disorders

NR

Quetiapine, olanzapine, lithium, risperidone, aripiprazole, ziprasidone, haloperidol

743

NR

During pregnancy

NR

Spontaneous abortion.

The increased risk of spontaneous miscarriage observed among women who were treated with antipsychotic medications during pregnancy was likely due to confounding factors.

Sakai T (2017) [78]

Cohort study

Psychiatric disorders

NR

Aripiprazole,

olanzapine,

quetiapine

165

NR

NR

NR

Miscarriage.

Aripiprazole may be associated with miscarriage. However, safety information regarding the use of aripiprazole during pregnancy is very limited. Further research is recommended.

Jacobson SJ (1992) [79]

Cohort study

Psychiatric disorders

Not exposed to teratogenic drugs

Lithium

286

EG (lit):

30 ± 5.3

NEG: 29.8 ± 5.3

First trimester

The postpartum follow-up was conducted when the average age was 61 weeks.

(1) Spontaneous abortion.

(2) Premature birth.

Lithium was not considered a significant human teratogen. Women with major affective disorders who wished to have children could continue lithium therapy, provided that adequate screening tests, including level II ultrasound and fetal echocardiography, were performed.

Newport DJ (2007) [6]

Cohort study

Psychiatric disorders

NR

Haloperidol,

olanzapine,

quetiapine,

risperidone

58

EG (hal):

30.5 ± 5.1

EG (ola):

30.8 ± 5.3

EG (que):

31.7 ± 6.8

EG (ris):

27.7 ± 4.4

During pregnancy proximate

Follow-up was conducted monthly for participants during pregnancy.

Preterm delivery.

All four antipsychotics (haloperidol, olanzapine, quetiapine, and risperidone) exhibited incomplete placental passage, with quetiapine showing the lowest placental passage among the medications studied.

Poels EMP (2022) [80]

Cohort study

Bipolar spectrum disorder

Unexposed with psychiatric disorders.

Lithium

99

NR

During pregnancy

NR

Premature birth.

There was no evidence of significant changes in neuropsychological function in children exposed to lithium in utero.

Ari Aripiprazole, EG Exposure group, Hal Haloperidol, Hal/Pen Haloperidol/Penfluridol, Lit Lithium, Lur Lurasidone, NEG Nonexposed group, NR Not reported, Ola Olanzapine, Que Quetiapine, Ris Risperidone 1FGAs refer to all first-generation antipsychotics in this study 2SGAs refer to all second-generation antipsychotics in this study 3Age was not reported for lamotrigine 4Only the mean value was presented

Correct Table 1

Table 1 Description of characteristics of included studies

Author, Year

Design

Indication

Nonexposed group

Interventions

Sample

Source

Age, years (mean ± SD/range)

Time of exposure

Duration of follow-up

Outcomes

Study conclusion

Huybrechts KF (2023) [46]

Cohort study

Psychiatric disorders

Unexposed with psychiatric disorders

Quetiapine, aripiprazole,

olanzapine, risperidone,

haloperidol

340,979

FGAs1: 28.54

SGAs2: 29.54

NEG: 27.54

First trimester

NR

Major congenital malformations.

(1) Overall, in utero exposure to antipsychotics was not significantly associated with an increased risk of malformations.

(2) Olanzapine may be associated with an increased risk of oral clefts, atypical antipsychotics with an increased risk of gastroschisis and brain anomalies, and chlorprothixene with an increased risk of cardiac malformations, but further research is needed.

Freeman MP (2021) [27]

Cohort study

Psychiatric disorders

Unexposed with psychiatric disorders

Aripiprazole

867

EG (ari): 32.4 ± 5.49

NEG: 32.7 ± 4.19

First trimester

The follow-up was conducted at 7 months of pregnancy and 12 weeks postpartum.

Major congenital malformations.

The risk of major malformations following early pregnancy exposure to aripiprazole was not significantly increased, but the findings are limited by the relatively small sample size. Larger studies are needed in the future to further validate these results.

Ellfolk M (2021) [26]

Cohort study

Psychiatric disorders

Unexposed with psychiatric disorders

Quetiapine, aripiprazole,

olanzapine,

risperidone,

clozapine

26,139

NR

First trimester

NR

Major congenital malformations.

Olanzapine use was associated with an increased risk of major congenital malformations and specifically, musculoskeletal malformations.

Cohen LS (2018) [47]

Cohort study

Psychiatric disorders

Unexposed with psychiatric disorders

Quetiapine

365

EG (que): 32.3 ± 4.89

NEG: 33.3 ± 4.07

First trimester

The follow-up was conducted at 7 months of pregnancy and 12 weeks postpartum.

Major congenital malformations.

Quetiapine was not a major teratogen. However, considering the uncertainty in risk estimation due to sample size limitations, the study results could only rule out approximately a fivefold increase in the risk of major malformations.

Bellet F (2015) [8]

Cohort study

Psychiatric disorders

Unexposed but unknown disease status

Aripiprazole

232

EG (ari): 31.8 ± 5.8

NEG: 31.4 ± 5.4

4 to 10 gestational weeks

The follow-up was conducted within 2 months after the expected delivery date.

(1) All congenital malformations (both major and minor).

(2) Preterm birth.

(3) Miscarriage.

No significant association was found between embryonic exposure to aripiprazole and major malformations. However, larger prospective studies are needed to further elucidate the reproductive safety of aripiprazole.

McKenna K (2005) [69]

Cohort study

Psychiatric disorders

Not exposed to teratogenic drugs, no mental illness or use of psychotropic drugs

Olanzapine

195

NR

First trimester

The follow-up was conducted 3 to 4 months after the expected delivery date.

Major congenital malformations.

Atypical antipsychotics did not appear to be associated with an increased risk of major malformations.

Diav-Citrin O (2005) [61]

Cohort study

Psychiatric disorders

Not exposed to teratogenic drugs

Haloperidol/penfluridol

709

EG (hal/pen): 32 (28–36)

NEG: 30 (27–33)

First trimester

The follow-up could extend up to 6 years, but most follow-ups were completed within the first 2 years after the infant’s birth.

(1) Major congenital malformations.

(2)Miscarriage.

(3) Preterm birth.

Pregnant women using haloperidol or penfluridol had a higher risk of preterm birth compared to the control group, but there were no significant differences in the risks of congenital malformations and miscarriage.

Patorno E (2017) [70]

Cohort study

Psychiatric disorders

Unexposed with psychiatric disorders

Lamotrigine,

lithium

1,325,563

EG (lit)3: 25.6 ± 6.1

NEG: 24.0 ± 5.8

First trimester

NR

All congenital malformations.

Lithium use by mothers in the first trimester of pregnancy was associated with an increased risk of cardiac malformations.

Cohen LS (2023) [46]

Cohort study

Psychiatric disorders

Unexposed with psychiatric disorders

Quetiapine,

lurasidone

1314

EG (lur): 33.2 ± 4.92

EG (que): 32.0 ± 4.80

NEG: 32.6 ± 4.18

First trimester

The follow-up was conducted at 7 months of pregnancy and 12 weeks postpartum.

Major congenital malformations.

Lurasidone and quetiapine did not appear to be major teratogens, but additional information is needed to refine the risk estimates.

Habermann (2013) [71]

Cohort study

Psychiatric disorders

NR

Aripiprazole, quetiapine, olanzapine, risperidone, clozapine, ziprasidone

447

SGAs2: 32 (27–36)

First trimester

The follow-up was conducted 8 weeks after the delivery date.

Major congenital malformations.

The study did not find that second-generation antipsychotics (SGAs) posed a significant teratogenic risk.

Kulkarni J (2014) [72]

Cohort study

Psychiatric disorders

NR

Clozapine,

olanzapine,

quetiapine,

risperidone

124

SGAs2: 32.67 ± 4.7

First trimester

The follow-up was conducted every 6–8 weeks during pregnancy.

Congenital malformations.

The rate of congenital anomalies in Australia was higher than the expected rate, particularly for congenital heart defects.

Viguera AC (2023) [73]

Cohort study

Psychiatric disorders

Unexposed with psychiatric disorders

Olanzapine

1207

EG (ola): 33.6 ± 4.56

NEG: 32.5 ± 4.08

First trimester

The follow-up was conducted at 7 months of pregnancy and 12 weeks postpartum.

Major congenital malformations.

Exposure to olanzapine in early pregnancy did not significantly increase the risk of major malformations, the study found.

Bodén R (2012) [74]

Cohort study

Bipolar disorder

NR

Lamotrigine,

lithium,

valproate,

carbamazepine

266

NR

During pregnancy

NR

Major congenital malformations.

Infants born to women with bipolar disorder had an increased risk of preterm birth, regardless of whether the mother received mood stabilizer treatment. Infants of untreated women also faced a higher risk of microcephaly and neonatal hypoglycemia.

Diav-Citrin O (2014) [75]

Cohort study

Psychiatric disorders

Unexposed with psychiatric disorders

Lithium

184

EG (lit): 32 ± 6

NEG: 31 ± 6

During pregnancy

The follow-up was conducted within the first 2 years after birth in most cases.

(1) Major congenital malformations.

(2) Miscarriage.

(3) Preterm delivery.

Lithium treatment during pregnancy was associated with a higher incidence of fetal cardiovascular abnormalities.

Källén B (1983) [76]

Cohort study

Psychiatric disorders

Exposed to psychotropic drug but unexposed to lithium with psychiatric disorders

Lithium

79

NR

First trimester

NR

Malformations.

The use of lithium in early pregnancy was found to possibly increase the risk of perinatal mortality and malformations in infants; therefore, it is recommended to avoid lithium during early pregnancy.

Vigod SN (2015) [9]

Cohort study

Psychiatric disorders

NR

Quetiapine,

olanzapine,

risperidone

834

SGAs2: 28.8 ± 6.1

NEG: 26.7 ± 6.3

During pregnancy

NR

Preterm birth.

Antipsychotic drug use in pregnancy had a minimal evident impact on important maternal medical and short-term perinatal outcomes. However, the rate of adverse outcomes is high enough to warrant a careful assessment of maternal and fetal well-being among women prescribed an antipsychotic drug in pregnancy.

Sørensen MJ (2015) [77]

Cohort study

Psychiatric disorders

NR

Quetiapine, olanzapine, lithium, risperidone, aripiprazole, ziprasidone, haloperidol

743

NR

During pregnancy

NR

Spontaneous abortion.

The increased risk of spontaneous miscarriage observed among women who were treated with antipsychotic medications during pregnancy was likely due to confounding factors.

Sakai T (2017) [78]

Cohort study

Psychiatric disorders

NR

Aripiprazole,

olanzapine,

quetiapine

165

NR

NR

NR

Miscarriage.

Aripiprazole may be associated with miscarriage. However, safety information regarding the use of aripiprazole during pregnancy is very limited. Further research is recommended.

Jacobson SJ (1992) [79]

Cohort study

Psychiatric disorders

Not exposed to teratogenic drugs

Lithium

286

EG (lit): 30 ± 5.3

NEG: 29.8 ± 5.3

First trimester

The postpartum follow-up was conducted when the average age was 61 weeks.

(1) Spontaneous abortion.

(2) Premature birth.

Lithium was not considered a significant human teratogen. Women with major affective disorders who wished to have children could continue lithium therapy, provided that adequate screening tests, including level II ultrasound and fetal echocardiography, were performed.

Newport DJ (2007) [6]

Cohort study

Psychiatric disorders

NR

Haloperidol,

olanzapine

quetiapine

risperidone

58

EG (hal): 30.5 ± 5.1

EG (ola): 30.8 ± 5.3

EG (que): 31.7 ± 6.8

EG (ris): 27.7 ± 4.4

During pregnancy proximate

Follow-up was conducted monthly for participants during pregnancy.

Preterm delivery.

All four antipsychotics (haloperidol, olanzapine, quetiapine, and risperidone) exhibited incomplete placental passage, with quetiapine showing the lowest placental passage among the medications studied.

Poels EMP (2022) [80]

Cohort study

Bipolar spectrum disorder

Unexposed with psychiatric disorders

Lithium

99

NR

During pregnancy

NR

Premature birth.

There was no evidence of significant changes in neuropsychological function in children exposed to lithium in utero.

Ari Aripiprazole, EG Exposure group, Hal Haloperidol, Hal/Pen Haloperidol/Penfluridol, Lit Lithium, Lur Lurasidone, NEG Nonexposed group, NR Not reported, Ola Olanzapine, Que Quetiapine, Ris Risperidone 1FGAs refer to all first-generation antipsychotics in this study 2SGAs refer to all second-generation antipsychotics in this study 3Age was not reported for lamotrigine 4Only the mean value was presented