• anxiety disorders
• panic disorder in pregnancy
• panic symptoms
• post-partum depression
• risk factors

Background. Anxiety symptoms are frequently reported by pregnant women and are often considered as part of the normal psychic experiences of pregnancy, especially if they are focused on the baby's health or on maternal competencies. Among all anxiety disorders, Panic Disorder (PD) needs to be carefully noted. It has been estimated that 3% to 12% of women experience symptoms related to PD at some time during their childbearing years, including during pregnancy and the postpartum period (Wenzel et al., 2001; Smith et al., 2004).
Although panic symptoms during the perinatal period are typical symptoms reported in the general population, they are often interpreted in the context of the perinatal state and the concerns about the pregnancy and fetus may represent the predominant features (Ross et al., 2006). As others have noted (Austin, 2003; Austin et al; 2005; Glover and O’Connor, 2002), it may be just as important to focus on the detection and treatment of perinatal anxiety, given its significant association with the development of subsequent Post-partum Depression (PPD). Although the role of anxiety disorders on the development of PPD has already been studied in literature, that of individual anxiety disorders has not received specific attention.
Aims. The aims of this thesis were: 1) to describe prevalence and clinical features of PD of a large non-clinical sample of women recruited at the 3rd month of pregnancy and to compare them with clinical features in a group of non-pregnant women of equivalent age 2) to investigate the role of PD (as family history, previous diagnosis or the occurrence of PD during pregnancy) on the development of PPD 3) to assess the specific role of PD in predicting probable depression(EPDS>12), minor or major depression (mMD) and false positives (EPDS>12 without mMD) at first month/year after delivery, compared to other anxiety diagnoses.
Methods. Participants: for the first aim, two samples of women with a diagnosis of PD were recruited; the first sample was composed of 43 pregnant women diagnosed with PD at 3th month of pregnancy, as part of a larger sample (N=1066) of a study conducted at Pisa by the Perinatal Depression-Research and Screening Unit (PND-ReScU). The second, the control group, was composed of 57 non-pregnant female outpatients diagnosed with PD, who presented to the psychiatric outpatient clinics of the same research center. For the second aim, the sample was composed of 600 pregnant women, as part of a larger sample (N=1066) of a study conducted at Pisa by the Perinatal Depression-Research and Screening Unit (PND-ReScU), who completed the assessment at 6th month after delivery; for the third aim, the sample was composed of 500 pregnant women who completed the assessment at the 12th month after delivery.
Instruments. A diagnosis of Axis I Panic Disorder was carried out with a Structured Clinical Interview for DSM-IV Axis I Disorders (SCID-I) (First et al., 1995) in the group of pregnant women, and with the Mini International Neuropsychiatric Interview (MINI) (Sheehan et al., 1998) in the control group of non-gravid women respectively. The clinical features of Panic Disorder were also evaluated with Panic-Agoraphobic Spectrum–Self-Report (PAS-SR) administered respectively at 6th month post-partum in the sample of pregnant women and at baseline in the control group. Symptoms of maternal depression were assessed using the 10-item Edinburgh Postnatal Depression Scale (Cox et al., 1987). The Post-partum Depression Predictors Inventory-Revised (PDPI-R) (Beck, 2002) was used to identify the risk factors for PPD. The family history of psychiatric disorders was assessed using the Family History Screen (FHS) (Weissman et al., 2000).
Statistical analysis. The Chi-square test was used to compare percentages and independent t-tests to compare mean scores. To assess the association between symptoms in the two groups Odds Ratios (Ors) with 95% confidence intervals were performed. Stepwise logistic regression models were performed to determine which panic symptoms were associated with PD in pregnancy. The alpha level was set at 0.05. Logistic regression models were used to estimate the association between PD, family history for PD and PPD. In order to establish the relationship between a specific anxiety disorder and the occurrence of probable depression, mMD and false positives in the postpartum, logistic regression models were performed. Odds Ratios and 95% of confidence intervals were reported. In order to clarify the magnitude of effect size we rescaled the OR in Cohen’s d using the γ coefficient (γ= (OR-1)/(OR+1)) (Kraemer and Kupfer, 2006). An effect size of 0.2, 0.5, and 0.8 are “small,” “medium,” and “large” (Cohen, 1988). Analyses were conducted using SPSS, version 15. Results. First research: One hundred and eighty four (17.3%) women had lifetime PD (life-PD), of whom 144 (13.3%) had a previous history of PD and 43 (4%) had current PD (curr-PD). Of the 43 women with current PD (28 (65.1%) with agoraphobia and 15 (34.9%) without agoraphobia), 5 (11.0%) had their first episode during the index pregnancy. As regards comorbidity, pregnant women with curr-PD (25.6%) were about six-fold more likely to have a major depressive episode (OR=5.844; 95% CI:1.513-22.563) than non-gravid women (5.6%). In the group of pregnant women, the symptoms most represented were: palpitations (90.7%), and shortness of breath (88.4%), choking (79.1%), sweating (74.4%). In the control group of non-pregnant women the symptoms most represented were: palpitations (87.7%), shortness of breath (63.2%), trembling (59.6%), sweating (57.9%). In a stepwise logistic regression a positive association was found between the presence of “fear of going crazy” (OR=4.020; 95% CI: 1.188-13.602), “shortness of breath” (OR= 9.970; 95% CI: 3.782-26.282) and panic disorder at the 3rd month of pregnancy. The presence of Agoraphobia was significantly higher in gravid women (OR 3.630; 95% CI: 1.557-8.462) than in non-gravid women. Pregnant women reported higher mean scores in the following manic factors of mood spectrum: psychomotor activation (3.923 vs. 2.563; t=2.705; p=0.008), creativity (3.461 vs. 1.618; t= 3.752; p= 0.000) euphoria (2.128 vs. 0.963; t=4.215; p<0.001). No significant differences were found in depressive factors scores in the two groups.
Second research: PD during pregnancy (RR=4.25; 95% CI: 1.48–12.19), a history of PD (RR 2.47; 95% CI: 1.11–5.49) and family history for PD (RR=2.1; 95% CI: 1.06–4.4) predicted PPD after adjusting for lifetime depression and risk factors for PPD.Third research: After adjustment for risk factors assessed with PDPI-R (RR=3.83; 95% CI: 1.84-7.97), a significant association was found between panic disorder (Adjusted RR=5.27; 95% CI: 2.0-13.91) and social phobia (Adjusted RR=3.80; 95% CI: 1.34-10.46) at the 3rd month of pregnancy and the likelihood of having probable depression at the first month postpartum. After the adjustment for risk factors assessed with PDPI-R (RR=2.74; 95% CI: 1.06-7.07), the predictive role of panic disorder (RR=7.23; 95% CI: 2.31-22.66) and social phobia (RR=6.63; 95% CI: 2.11-20.85) at the 3rd month of pregnancy remained significant in predicting 1st month postpartum mMD. With regard to the one-year period prevalence, our results show that after adjustment for the established risk factors assessed with PDPI-R (RR=4.66; 95% CI: 2.41-9.02), the only anxiety disorder associated with postpartum depression was panic disorder (Adjusted RR=3.10; 95% CI: 1.16-8.22). OCD (Adjusted RR=8.66; 95% CI: 1.59-47.03) was associated with false positive cases assessed at the 1st month postpartum, while no anxiety disorders were further associated with false positives assessed during the 1st year after childbirth. Limitations:First, the response rate is moderately low (49.9%), however it is comparable to that of similar studies that used self-report and structured clinical interviews, in which the observation period spanned from pregnancy to post-partum (Grant et al., 2008; Kitamura et al., 2006). The socio-demographic characteristics of women who refused to participate in the study are not available because data collection was possible only after the informed consent form was signed, as prescribed by the Italian law on privacy; thus we were unable to ascertain if non-responders were mostly of lower socio-economic class and had lower levels of education. Finally demographic characteristics of the two sample (pregnant women and control group) were not homogeneous: pregnant women had higher education levels compared to the control group and were more frequently married or living with the partner.
Conclusions. Symptoms of panic during pregnancy are typical of panic symptoms of the general population. But symptoms such as “being afraid of going crazy” and “shortness of breath” are significantly higher in pregnant women with PD compared to the control group, suggesting that these symptoms, particularly “fear of going crazy”, may be discriminating for discerning panic during pregnancy from panic manifestation of other periods of life. Furthermore, PD (family history, current PD, history PD) represents an important risk factor for the development of PPD and should be routinely screened in order to develop specific preventive interventions. It is important to underline that the role of a specific anxiety diagnosis in predicting depressive symptoms is related both to outcome definitions and to time of assessment. PD predicted both 1st month PPD than probable depressions. In relation to the 12th month postpartum, the results of this study confirmed the role of Panic Disorder in predicting a diagnoses of minor or major depression even if with a decrease in effect size.
Carrying out antenatal screening of established risk factors and accurate diagnoses of Panic Disorder during pregnancy may help to plan adequate treatment in order to prevent possible postpartum distress outcomes.