Delusional Misidentification Syndromes in Postpartum Psychosis: A Systematic Review

Delusional misidentification syndromes (DMS) are a group of rare psychopathological experiences that occur in psychosis and are characterized by the misidentification of a person or place. They include Capgras syndrome, in which a familiar person or object is believed to have been replaced by a duplicate or imposter [1]; Fregoli syndrome, where strangers are believed to be a familiar person in disguise [2]; and other syndromes such as intermetamorphosis, the syndrome of subjective doubles, reduplicative paramnesia, delusional companions, and mirrored self misidentification [3].

Initially, it was thought that DMS arose due to a combination of psychotic illness and psychological factors [4]. However, it is now recognized that there may be a structural basis for DMS involving disruption in the right hemisphere brain regions responsible for facial processing and recognition [5]. Estimates of the prevalence of DMS in inpatient psychiatric populations vary from 1.3 to 4.1% [6], and they are generally believed to be uncommon in perinatal mental illness [7].

Mental health problems affect up to 20% of women in the perinatal period, and biological factors play a significant role in the development of these conditions [8]. One of the most serious perinatal mental disorders is postpartum psychosis, which occurs in 1–2/1,000 women and has significant risks to the wellbeing of the mother and infant including suicide and infanticide [9]. In the early phase of illness, many women report difficulty in caring for and bonding with the baby [10]. In the later stages, women can develop delusions involving the infant, including that the baby may have an altered identity or be a “changeling” [11]. Previous research has shown that these mothers are more likely to inflict abuse towards the infant and be considered unsafe caregivers than mothers with delusions of persecution towards the infant [12].

DMS are often accompanied by hostility towards the object of delusional misidentification [13], and DMS in postpartum psychosis is a particular concern due to the risk of neglect, violence, or infanticide towards a misidentified child [14]. This review aimed to collate all published cases of DMS in postpartum psychosis to further understand how they can present in perinatal mental illness, how they affect the mother-infant bond including risk to the infant, and to consider future scope for research.

An online search was conducted in August 2021 using the following databases: PubMed, MEDLINE, PsycINFO, CINAHL, and Embase. The following search was used: “Delusional misidentification OR Fregoli OR Capgras OR Intermetamorphosis OR Subjective doubles OR Mirrored self misidentification OR Delusional companion OR Reduplicative paramnesia,” combined by “AND” with the search “Perinatal OR Postnatal OR Birth OR Postpartum OR Puerperal OR Prenatal OR Puerperium.”

Papers were eligible for inclusion if they described DMS in the perinatal period, which was defined as during pregnancy or up to 1 year following the birth of a child. Papers were not excluded on the basis of study type, language, or publication date.

Papers identified by the online search were screened by title and abstract to identify duplicates. The full texts of unique papers were reviewed to determine if they fit the inclusion criteria. Reference lists of included papers were checked for further relevant papers.

The quality of each paper was assessed. No papers were excluded on the basis of low quality. Case reports and Case Series were appraised using the Joanna Briggs Institute (JBI) Critical Appraisal Checklist for Case Reports [15] by two independent raters (authors G.L. and L.B.), and discrepancies in scoring were resolved through discussion to reach a consensus score.

The following data was collected and entered into a Microsoft Excel database: title, publication year, country of publication, patient age, psychiatric history, medical history, obstetric history, family history, social circumstances, investigations, psychological factors, treatments, onset of DMS, duration of DMS, subject of delusional misidentification, description of DMS, accompanying psychiatric symptoms, content of accompanying delusional beliefs, documentation of cognition, and effect of DMS on mother-infant bond.

A total of 55 references were identified by the online search. After the removal of duplicates 20 papers remained. One paper was excluded on the basis that the full text could not be sourced for further review [16], and 8 met the inclusion criteria for the review. Searching of references identified 1 further paper, which met the criteria, giving 9 papers in total (shown in Fig. 1).

Eight of the included papers were single case reports describing Capgras syndrome in the perinatal period [7, 17‒23]. The ninth was a case series of 4 cases of Fregoli syndrome identified by a single Mother and Baby Unit [24]. No research studies or literature reviews were found. Cases had been reported internationally, although the majority were from Europe, and publication dates ranged from 1971 to 2021.

Quality appraisal using the JBI Critical Appraisal Checklist for Case Reports showed high levels of initial discrepancy between raters, with a percentage agreement of 42% from independent rating. Consensus scored varied from 3 to 8 out of 8 (summarized in Table 1).

Patients described by the reports varied from 26 to 36 years old and in 8 of the 12 cases reported were primiparous. Three of the reports specified the mode of delivery and all of these were by caesarean section (CS). None of the cases comment on if the mother and infant were separated in the immediate postnatal period. Six of the twelve cases had no history of mental illness, while in the other six the most common history was of postpartum (or post miscarriage in one case) psychosis (n = 3), and other cases reported history of depression (n = 2), and postnatal depression (n = 1). One case reported a history of subarachnoid haemorrhage with subsequent personality change. Eight cases commented specifically on family history of mental illness, of which one identified a family history of bipolar disorder and another of perinatal mental illness (summarized in Table 2).

Aetiological investigations varied between studies, from no reported investigations to a thorough medical workup including blood tests, CT head, and EEG. Three cases identified organic pathology that may have contributed to the presentation: 1 patient had pre-eclampsia and severe postnatal hypertension and nephrotic syndrome, and two had neurological abnormalities – ventricular enlargement after previous subarachnoid haemorrhage, and ventricular enlargement with right temporal atrophy. A further 3 case reports considered psychological origins of the delusions, and the series of 4 cases of Fregoli Syndrome suggested ambivalence arising in childbirth as a possible contributor to pathogenesis.

Most patients received antipsychotic treatment, either alone (n = 5) or in combination with ECT (n = 3). Two cases received ECT alone, and two had no psychotropic treatment reported (summarized in Table 3).

In 6 cases, there were multiple subjects of misidentification, while in the other 6, only a single subject was misidentified. The most common subject of misidentification was the patient’s husband (n = 7), followed by their baby (n = 6), hospital staff (n = 4), other family members (n = 3), and self (n = 1). In the 4 cases of Fregoli syndrome, the misidentified person(s) were identified as the patient’s family members. In 10 cases, there was no documentation of justification given for the delusional beliefs. One case cited differences in physical appearance and one behavioural change as the reason for the beliefs. The duration of DMS reported varied from 2 days to 5 years, with the majority of cases lasting days to months. Ten of the twelve cases were diagnosed with psychosis involving DMS within 3 weeks of delivery and one at 2 months postpartum. Another had psychosis since giving birth, which developed into DMS at 9 months postpartum (summarized in Table 4).

A wide range of presentations were reported including schizophreniform illness with thought insertion and passivity, depressive symptoms such as low mood and guilt, and manic features such as flight of ideas and grandiosity. Accompanying delusional beliefs include delusions of paranoia (n = 3), guilt (n = 1), grandiose/religious (n = 1), and Ekbom’s syndrome of delusional infestation (n = 1). Four cases also had auditory hallucinations, one of which was related to her DMS, while others were paranoid or guilt-related. No cases reported dissociative features such as depersonalization or derealization.

Five of the cases specifically remark on the impact of perinatal illness on the maternal-infant bond, of which four result in the mother’s being unwilling to care for the infant due to their delusional beliefs, including one case where “This belief was so firmly held that she could not bring herself to feed the child,” and one where “she became withdrawn, disinterested, expressed persecutory ideas, and neglected her baby.” In one case, the mother believes the infant is not her child but is happy to continue caring for them (summarized in Table 5).

The results of this review show that although DMS are believed to be rare in perinatal mental illness, there has been a small but steady trickle of cases reported since the 1970s. Reliably assessing the quality of case reports proved challenging, possibly reflecting the limitations of available scoring systems, and the heterogeneous nature of case reports identified in this review, many of which were produced prior to the introduction of standardized guidelines for case reporting [25].

Attempts to identify cases of DMS in postpartum psychosis are complicated by the fact that there is no unique ICD-10 code for postpartum psychosis, so patients with postpartum psychosis may not be identified in studies which look only at certain aspects of the clinical record rather than reviewing the case as a whole [26]. A review of 255 published cases of Capgras syndrome conducted in 2019 counted puerperal psychosis in the “Other” diagnostic category, which consisted of a total of 11 cases, including 8 that had no diagnosis specified in the original report. Although no further breakdown is given of how many of the remaining 3 cases identified in the review were patients with puerperal psychosis, our review identified 7 case reports of Capgras syndrome in the perinatal period prior to 2019, showing that without a highly robust search strategy many of these cases are easily missed [27].

The spectrum of psychiatric symptoms in the cases identified by this review varied widely, reflecting the clinically heterogenous nature of postpartum psychosis, which can resemble affective disorders, schizophreniform illness, and organic psychosis [28]. That DMS was identified throughout the range of postpartum psychosis presentations, and in the presence and absence of physical health abnormalities, obstetric difficulties, and psychological precipitants, also indicates the varied aetiological origin of DMS. Although the sample size is small and there are few immediately striking commonalities between the reports, it is of note that 2 of the 12 identified cases had evidence of right hemisphere brain dysfunction, which is known to be associated with the development of DMS and is not a common finding in postpartum psychosis as a whole [29]. Women with preexisting right hemisphere dysfunction or who are identified as having this on brain imaging may, therefore, be at an increased risk of developing DMS during an episode of postpartum psychosis and merit close clinical observation. The treatment given varied greatly between reports, likely reflecting the broad geographical and temporal spread of included cases.

In the 8 cases of Capgras Syndrome identified in this review, the subject of misidentification tended to be people of emotional significance to the patient, most commonly the partner or infant, and in those who misidentified the infant, this often led to them being unable or unwilling to provide safe care. This is similar to the findings for Capgras Syndrome as a whole [27], but is particularly relevant for a perinatal population where the impact of such delusions may damage the developing mother-infant bond, wider family relationships, or may even present risk to the wellbeing and life of the infant [14].

The literature search did not identify any previous studies or reviews focussing on this area, making it difficult to estimate the true prevalence of these conditions in postpartum psychosis. However, a 2006 study investigating the content of delusions in 105 mothers with postpartum psychosis found that many had experiences indicative of DMS (21.5% of interviewed mothers believed the baby belonged to someone else, 36% that the baby was the devil (or ill fated), 12.5% that the baby was God, and one that the baby was a dead relative) [12]. This indicates that if actively explored, DMS may be far more common in postpartum psychosis than previously believed, a view echoed by some of the source research accounts [18]. Future studies of this clinical population should therefore begin by establishing the prevalence of DMS in patients with PPP and further investigating the risk these delusions pose to the mother’s mental health as well as the wellbeing of the infant and development of the mother-infant bond.

Future research could also identify a more systemic evaluation of aetiology, including whether there are any transcultural issues related to the DMS, or if their aetiology was in any way linked to historical trauma. Further psychological treatments that allow women to understand the DMS in a timely manner may also support earlier recovery from the DMS.

The published literature on DMS in postpartum psychosis is confined to a small number of case reports and a single case series. This is the first attempt to collate the knowledge in this area through systematic review. Although small in number, these cases reveal several important learning points including that DMS can occur throughout the spectrum of postpartum psychosis presentations, with or without underlying organic disease, and may be particularly associated with right hemisphere brain dysfunction. Active exploration of the nature of delusions in mothers with postpartum psychosis is required to identify cases of DMS to mitigate the risk of harm to the infant and mother-infant bond and may also uncover that these syndromes are more common in postpartum psychosis than currently believed.

We would like to express our deep gratitude to Professor Femi Oyebode for his invaluable encouragement and guidance during the writing of this paper.

An ethics statement in not applicable because this study is based exclusively on published literature.

The authors have no conflicts of interest to declare.

The authors did not receive any funding.

The initial conception was provided by Gertrude Seneviratne and first draft was provided by Gabriella Lewis. Subsequent rewriting of the text was done by Gabriella Lewis, Lucy Blake, and Gertrude Seneviratne.

Data sharing is not applicable to this article as no new data were created or analysed in this systematic review.