Learning opportunities in eclampsia care: A ten‐year audit from Norway, 2013–2022 Authors: Camilla Tjønneland Mentzoni, Kjerstine Røe, Solveig Bjellmo, Marte Irene Skille Carlsen, Marit Halonen Christiansen, Tora Sandvik Fjeldstad, Lena Henriksen, Truc Trung Nguyen, Lill Trine Nyfløt, Leiv Arne Rosseland, Chen Sun, Oddvar Uleberg, Siri Vangen, Kari Klungsøyr, Hilde Marie Engjom, OCINOR—Eclampsia Working Group Published: 2026 ABSTRACT AbstractIntroductionThe World Health Organization recommends surveillance, including audits with case reviews of severe maternal morbidities such as eclampsia, to improve obstetric outcomes. We aimed to audit eclampsia and to identify learning opportunities in eclampsia care in Norway.Material and MethodsOur study population included all women discharged with a diagnosis of eclampsia (ICD‐10: O15) and with a documented generalized seizure in medical records at two university hospitals in Norway from 2013 to 2022. An eclampsia working group developed a digital eclampsia case report form for gathering clinical information from medical records. Based on a structured summary of information from the digital case report form, the group reviewed the care for women with eclampsia. Each case was discussed and reviewed according to the clinical guidelines by the group. Learning opportunities to improve care were identified and classified according to whether they contributed to the outcome.ResultsAmong 93 139 deliveries, 22 women with an initial eclampsia diagnosis were identified (2.4 per 10 000). Subsequent clinical information identified an alternative probable etiology of seizures in five women. None of the 22 women received magnesium sulfate prior to the seizure, and cerebral imaging was performed in 11 of 22 women. In 18 women, learning opportunities were identified, and for eight women, a different management could potentially have prevented eclampsia. The eclampsia working group identified learning opportunities related to underuse of magnesium sulfate prophylaxis when indicated, inadequate management of hypertension both before and after eclamptic seizure, underuse of cerebral imaging as a diagnostic tool, and lack of follow‐up consultation with an obstetrician in women diagnosed with eclampsia.ConclusionsLearning opportunities were found in 18 of 22 women, and improvements in care could potentially have changed the outcome for eight of the women. This audit highlights the importance of differential diagnostic consideration for seizures in pregnant and postpartum women, the use of prophylactic magnesium sulfate when indicated, and the diagnosis and management of hypertension in pregnancy. Eclampsia audits should be incorporated into national routine surveillance and learning systems to improve obstetric care.Keywords: clinical care, eclampsia, hypertensive disorder of pregnancy, obstetric care, severe maternal morbidity Abbreviations confidence intervals case report form computer tomography international classification of diseases magnesium sulfate magnetic resonance imaging Key message This audit of eclampsia resulted in learning opportunities. Improved hypertension management and prophylactic magnesium sulfate represent opportunities for prevention, while cerebral imaging and follow‐up postpartum were underused. Clinicians need to consider differential diagnosis in pregnant and postpartum women with seizures. INTRODUCTION The World Health Organization recommends surveillance and learning from severe maternal and perinatal outcomes in a continuous effort to improve health care during pregnancy and childbirth. 1 Learning from severe maternal morbidity is an essential component in the improvement of obstetric care. 1 , 2 The Norwegian maternal mortality group has reviewed all known cases of maternal deaths from 1996 to the present as part of a research project. 3 There are currently no routine audits with confidential inquiries into severe maternal morbidity in Norway. 4 Audit of cases of severe maternal morbidity through confidential inquiries adds a comprehensive clinical perspective to traditional register‐based surveillance. 5 While studies on individual maternal risk factors for severe maternal morbidity are important, audits can identify variations in management that influence morbidity risk at the group level, generate generalizable lessons to improve care, and uncover system‐level issues that need to be addressed. 5 , 6 Furthermore, examining the treatment course of the women who survived can give valuable insights into the care provided. 5 , 6 , 7 Eclampsia is a severe maternal morbidity defined by generalized tonic–clonic seizures related to pregnancy, accompanied by signs of preeclampsia and exclusion of other causes of seizures. 8 Key elements of good clinical care to prevent and treat eclampsia include management of hypertension and the use of magnesium sulfate (MgSO4). 8 , 9 Audits of eclampsia in other countries have identified important learning points to improve obstetric care. 10 A previous Scandinavian audit of eclampsia cases from 1998 to 2000 demonstrated substandard care to 42% of women with eclampsia. 11 The aim of this study was to audit the health care for women with eclampsia and identify learning opportunities to improve future obstetric health care. MATERIAL AND METHODS Source and study population The source population for the study was all women giving birth at Haukeland and Stavanger University Hospitals in Western Norway from 2013 to 2022 (N:93139). 12 Women with eclampsia were identified using the ICD‐10 code O15 (eclampsia) in the hospital database, and clinical information was retrieved from medical records by clinicians at the collaborating institutions using a digital case report form (CRF). Among the identified cases with O15 eclampsia, only women with generalized seizures documented in medical records were included. Audit process An eclampsia working group was established to develop the CRF used to collect clinical information, to participate in data collection, and to review the health care provided (Figure 1). The working group included medical professionals within different specialties (obstetrics, anesthesiology, and general practice) and user representatives from the Norwegian Association of Midwives, the Norwegian Society of Gynecology and Obstetrics, and the Norwegian Women's Public Health Association. Overview of the audit process. The digital CRF was developed for the standardized collection of clinical information (Figure 1). The case report form by the United Kingdom Obstetric Surveillance System (UKOSS) served as a template and was supplemented by key information listed in the guidelines for diagnosis and management of hypertensive disorders in pregnancy issued by the Norwegian Society of Obstetrics and Gynecology. 8 , 13 , 14 , 15 The CRF was developed through discussion within the eclampsia working group, resulting in a consensus‐based agreement on key information and clinical variables. The CRF covered maternal characteristics, information about symptoms, signs, and aspects of care throughout the clinical management of eclampsia. The digital CRF included both general and optional information, with a maximum of 264 questions across nine main themes, resulting in up to 302 variables for analysis (Table 1). Examples of optional questions included the prehospital management for women with the first seizure outside of the hospital or information about perioperative care in cesarean births. At the collaborating hospitals, clinical information from medical records was retrieved and entered into the digital CRF. Most responses had predefined options or required numerical input, while a subset permitted free‐text responses. Upon submission, the de‐identified CRF was automatically transferred and stored on a secure research server for further analyses. A secure web‐based interface was developed to summarize, structure, and present the information gathered from the CRF for expert review. TABLE 1Main themes in the case report form used for audit.Themes for the case report form1Background, including social vulnerability2Pregestational medical risk factors3Follow‐up during pregnancy (general and related to hypertensive pregnancy disorder)4Eclampsia—event‐specific information, initial response, and prehospital treatment5Diagnostic information and monitoring of blood pressure6Delivery‐related information, including fetal surveillance and anesthesiology in case of C‐section7Medical treatment, intensive care, complications, and hospital stay8Follow‐up after discharge9Delay in care and department routines Main themes in the case report form used for audit. The eclampsia working group reviewed the care for women with eclampsia in plenary discussions. The review was based on Norwegian clinical guidelines valid at the time of the relevant pregnancy and reviewers' clinical experience. 8 , 13 , 14 Building on the classification used for maternal confidential inquiries in the UK, we used the following categories to classify learning opportunities: 0—good care, 1—improvements in care which would make no difference to the outcome, 2—factors that could have contributed to increased severity of the outcome, 3—different management could have prevented the severe outcome, 4—not enough information to review. 16 Delays in receiving timely and appropriate care were also evaluated, both by the reporting clinician and in plenary discussions. Delays were classified as patient‐related delay, delay in reaching an adequate health facility, and provider‐related delay. 7 , 17 Statistical analyses In addition, a quantitative analysis of the data was performed with frequency measures and cross‐tabulation. On selected variables, 95% confidence intervals (CI) were estimated. The analyses were performed using STATA 19 (StataCorp LLC, College Station, TX, USA). RESULTS Characteristics of the study population Among 93 139 women who gave birth at the two university hospitals during the study period, 22 were diagnosed with eclampsia, corresponding to an incidence of 2.4 per 10 000 births. Maternal characteristics are presented in Table 2. The mean age was 30.1 years, and 68% (15/22) were nulliparous. Most women were born in Norway (16/22). One woman had a multifetal gestation. Cesarean section was the mode of delivery in half of the women (11/22). Among women with antepartum seizures, 8/10 delivered by cesarean section, compared to 1/8 of postpartum cases. TABLE 2Characteristics of study population (N:22). N (%;/IQR)Maternal age (mean)30.1 (27–33)ParityPara 015 (68)Para≥17 (32)Mother's country of birthNorway16 (73)Other6 (27)Body mass index (kg/m2) at start of pregnancy<3017 (77)≥305 (23)Mode of deliverySpontaneous vaginal8 (36)Assisted vaginal3 (14)Cesarean section11 (50)Emergency cesarean section9 (82)Elective cesarean section2 (18)Gestational age at birth (weeks)≤34 weeks8 (36)340–366 weeks2 (9)≥37 weeks12 (55) Characteristics of study population (N:22). Most women (18/22) had their first seizure in the hospital, and the remaining four women had their first seizure at home (Table 3). Seizures occurred antepartum in 10 women, intrapartum in four, and postpartum in eight women. Among postpartum cases, half (4/8) occurred in the labor and delivery unit, indicating a short interval between delivery and seizure. TABLE 3Setting, signs, symptoms, treatment and follow‐up in the study population (N:22), Stavanger and Haukeland University Hospital, 2013–2022, Norway.EclampisaOther diagnosis likely a N (%) N (%)Total175SettingHome4 (24)Obstetric observation unit2 (12)Labor and delivery unit7 (41)4 (80)Postpartum unit2 (12)1 (20)High dependency unit2 (12)TimingAntepartum9 (53)1 (20)Intrapartum2 (12)2 (40)Postpartum6 (35)2 (40)Known preeclampsia b 9 (53)0Symptoms/signs of preeclampsia/imminent eclampsiaSymptoms prior to seizure c Quivering1 (6)0Hyperreflexia2 (12)0Headache10 (59)0Upper abdominal pain2 (12)0Nausea2 (12)1 (20)Visual disturbance6 (35)0Shortness of breath00One or more of the symptoms above11 (65)1 (20)Classic symptoms: headache and/or visual disturbance11 (65)0Proteinuria reported in primary care or at hospital admissionProteinuria ≥1 + on urine dipstick15 (88)2 (40)Proteinuria ≥3+ on urine dipstick10 (59)0 Missing information on proteinuria 1 (6)1 (20)At least one symptom or presence of proteinuria ≥3+13 (76)1 (20)Abnormal lab values13 (76)2 (40)Hemolysis (haptoglobin <0.2 g/L)3 (22)0 Missing haptoglobin information 3 (18)3 (60)Thrombocytopenia <100 × 109/L2 (12)0 Missing thrombocytopenia information 1 (6)0Elevated ASAT/ALAT d 10 (59)2 (40)Creatinine >90 μmol/L or doubled from the previous value7 (41)0Multiple seizures2 (12)3 (60)Magnesium sulfateMagnesium sulfate as prophylaxis00Magnesium sulfate as treatment17 (100)5 (100)Duration of magnesium sulfate treatment≥24 h14 (82)5 (100)12–23 h1 (6)0<12 h1 (6)0 Missing information on duration 1 (6)0Monitoring blood pressure prior to seizureWithin the last 24 h prior to general seizure (Yes)13 (76)4 (80)No measurement the last 24 h prior to the general seizure4 e (24)1 (20)Within the last hour prior to seizure (Yes)6 (35)2 (40)Antihypertensive treatmentAt any time17 (100)2 (40)Prior to and after seizures7 (41)0After seizures only10 (59)2 (40)Route of administration of antihypertensive medicationOral6 (35)0Intravenous2 (12)0Both oral and intravenous9 (53)2 (40)Cerebral imaging (CT or MRI)8 (47)3 (60)Follow‐up (planned or documented)With general practitioner13 (76)4 (80)With obstetrician8 (47)1 (20)Abbreviations: CT, computer tomography; MRI, magnetic resonance imaging. a In five women, subsequent medical information indicated another more likely cause of the generalized seizure. b Preeclampsia was defined as repeatedly measured blood pressure >140/90 mmHg after 20 weeks of gestation and proteinuria with ≥1+ on urine dipstick or urine protein ≥0.3 g/24 h in the national obstetric guideline during our study period. c One or more symptoms may be reported for each woman. d Value above reference level at the hospitals. e All of the four cases occurring at home. Setting, signs, symptoms, treatment and follow‐up in the study population (N:22), Stavanger and Haukeland University Hospital, 2013–2022, Norway. Abbreviations: CT, computer tomography; MRI, magnetic resonance imaging. In five women, subsequent medical information indicated another more likely cause of the generalized seizure. Preeclampsia was defined as repeatedly measured blood pressure >140/90 mmHg after 20 weeks of gestation and proteinuria with ≥1+ on urine dipstick or urine protein ≥0.3 g/24 h in the national obstetric guideline during our study period. One or more symptoms may be reported for each woman. Value above reference level at the hospitals. All of the four cases occurring at home. All 22 women were initially diagnosed and treated as having eclampsia at the time of the generalized seizure, and all were discharged with a diagnosis of eclampsia (ICD‐10: O15). Subsequent review of the medical records identified alternative, more likely causes in five women (Table 3), reducing the confirmed cases to 17. Epilepsy accounted for three of the re‐classified cases, where all women subsequently received antiepileptic medication. One woman, who was subsequently diagnosed with epilepsy, was treated for eclampsia in two pregnancies (the last was the index pregnancy in our study). In the two remaining women, the most likely causes of seizures were considered to be psychogenic non‐epileptic seizures and hyponatremia, respectively. Signs and symptoms As shown in Table 3, the signs and symptoms differed among the 17 women with eclampsia and the five women with a likely alternative cause of seizures. Signs and symptoms of preeclampsia were more common in women with eclampsia than in women with an alternative diagnosis, and 53% (9/17) had a known diagnosis of preeclampsia prior to the eclamptic seizure. At least one symptom related to preeclampsia or imminent eclampsia was reported prior to the seizure in 65% (11/17) of women with eclampsia, most frequently headache (59%, 10/17) and visual disturbance (35%, 6/17). Further, two of the six asymptomatic women were treated with antihypertensive medication before the eclamptic seizure. Among the women with non‐eclamptic seizures, only one had any preceding symptoms (nausea). Proteinuria (≥1+ on urine dipstick) was reported in primary care or at hospital admission in 88% (15/17) of eclampsia cases and 2/5 non‐eclamptic cases. Proteinuria ≥3+ on urine dipstick was only reported in eclampsia cases (59%; 10/17). Abnormal laboratory values were reported in 77% (13/17) of eclampsia cases and in 40% (2/5) of non‐eclamptic cases. Multiple seizures were reported in 12% (2/17) of the eclampsia cases and in 60% (3/5) of the non‐eclamptic cases. As Table 3 displays, 13 of 17 cases of eclampsia occurred in the hospital, and seven of these at the labor and delivery unit. All the women who had antepartum eclampsia in the hospital had a known diagnosis of preeclampsia (5/5). Among the eight women with intra‐ and postpartum eclampsia, three had a known diagnosis of preeclampsia prior to the seizure. All eclamptic seizures were reported within the first day postpartum. Among the women with non‐eclamptic seizures, one had antepartum seizures, two had intrapartum, and two had postpartum seizures (Table 3). Among the three women who were subsequently diagnosed with epilepsy, all seizures occurred intra‐ or postpartum, the last occurring on day two postpartum. Treatment and follow‐up None of the 22 women diagnosed with eclamptic seizures received prophylactic MgSO4 (Table 3). All received MgSO4 as treatment after the initial seizure. Among the 18 cases who had seizures in hospital, MgSO4 treatment was started after an average of 9.4 min (95%CI: 5.98–12.80). In two of the four women with the first seizure at home, MgSO4 was administered by prehospital services. Treatment with MgSO4 was continued for 24 h in 19/22 (86%) women. One woman received benzodiazepine as the initial treatment. All 17 women with verified eclampsia received antihypertensive treatment; 41% (7/17) received antihypertensive treatment before the eclamptic seizure, and all received treatment afterward. The route of administration was oral in 35% (6/17) and intravenous (with or without oral) in 65% (11/17) (Table 3). Among the five women with non‐eclamptic seizures, none received antihypertensive medication before the seizure, and two received antihypertensive medication after the seizure. Blood pressure was monitored within the last hour prior to the seizure in 35% (6/17) of eclampsia cases. In the remaining seven women with eclampsia occurring in the hospital, blood pressure was monitored within the last 24 h. In the non‐eclamptic group, only one woman did not have her blood pressure measured within 24 h prior to the seizure. Cerebral imaging, by computer tomography (CT) or magnetic resonance imaging (MRI), was performed in 50% (11/22) of the total number of women with seizures. Cerebral imaging was also omitted in two women (2/5; 40%) with multiple seizures, both of whom were in the non‐eclamptic group. Follow‐up after discharge was planned with the general practitioner for 77% (17/22) of the women, while 41% (9/22) had a follow‐up consultation with an obstetrician. Complications were observed in three women with eclampsia: stroke (1/17), hypertensive retinopathy (1/17), and pulmonary edema (1/17). Two cases of intrauterine fetal death were reported, both among the 17 women with eclampsia (12%), and both occurring in women with antepartum eclampsia in early gestation before week 28. Care review and learning opportunities Figure 2 shows the learning opportunities identified by the eclampsia working group. There was sufficient information to assess the care for all the women. When assessing the quality of care, opportunities for improvements were identified in 82% (18/22); and in 36% (8/22), a different management could potentially have changed the outcome (Figure 3). The most frequent learning points identified during the review were related to the potential use of MgSO4 as prophylaxis and as prehospital treatment; need for improved monitoring and treatment of hypertension; underuse of cerebral imaging, and underuse of postpartum follow‐up consultation with an obstetrician (Figure 2). In addition, improvements in pre‐pregnancy care were identified as a learning point. Learning points identified among the study population (N:22). Classification of care based on identified learning opportunities and whether improvements in care could have changed the outcome, in the study population (N:22). Using the four categories to classify learning opportunities to improve care, four women were classified as category 0, and the overall assessment of their case reports indicated that they had received good care (Figure 3). Ten women were classified as category 1, where learning opportunities that probably would not have altered the outcome were noted. This was primarily related to a lack of cerebral imaging (6/10). Eight women were classified as category 3, where a different management could potentially have prevented eclampsia. This was related to not giving prophylactic MgSO4 (5/8) and the management of hypertension (3/8). Among these eight women, seven had established preeclampsia, and six had preterm eclampsia. Seven women had eclampsia in the hospital, all of whom reported neurological symptoms. Two women were in a high‐dependency unit when eclampsia occurred. Antihypertensive treatment had been initiated prior to eclampsia in all women for whom the use of prophylactic MgSO4 was reviewed as a learning point. In the three women where improved management of hypertension was a learning point, two had a diagnosis of preeclampsia prior to eclampsia. In one case occurring at home, elevated blood pressure (160/98–170/100 mmHg) without further treatment was reported two days prior to eclampsia. For one woman, high systolic values (>160 mmHg) and headache were reported during labor, without adequate treatment of hypertension and eclampsia occurred 30 min postpartum. For one woman, antihypertensive treatment was delayed for two hours after eclampsia despite systolic blood pressure in the range 170–189 mmHg. When reviewing delays in treatment, patient factors contributed to the delay in one case, as recommended management for a high‐risk pregnancy was not followed. Delay in reaching the appropriate hospital department caused delayed treatment with MgSO4 for one woman presenting with a seizure in pregnancy, with initial admission to a general medical unit followed by transfer to the obstetric department. Potential provider‐related delays were reported in six cases related to clinical assessment, acknowledging and treating hypertension. DISCUSSION This audit of clinical care over a ten‐year period for women diagnosed with eclampsia provided valuable insight into the diagnosis and management of eclampsia and identified several learning points. For verified eclampsia cases, we identified an underuse of prophylactic MgSO4 and delays in the detection and treatment of hypertension. The underuse of cerebral imaging was evident in both women with an eclamptic and a non‐eclamptic seizure. Importantly, the audit process revealed that 23% (5/22) of the women meeting the inclusion criteria (generalized seizures and eclampsia as the discharge diagnosis) had a more likely alternative diagnosis. This highlights the need for a broad diagnostic approach following generalized seizures, also in pregnant and postpartum women. Maternal characteristics in the study population were comparable to the source population with regard to maternal age and proportion of Norwegian‐born mothers, but with a higher proportion of nulliparous women and multifetal gestations, both of which are known risk factors for eclampsia. 8 , 12 A positive finding in our audit was the quick, timely, and adequate treatment with MgSO4 when the generalized seizures occurred in an obstetric hospital department. Overall, 95% of cases received MgSO4 as first‐line treatment, which should be near 100% according to the WHO. 1 We assume this is due to well‐established guidelines and protocols concerning emergency obstetric events, such as eclampsia. 8 Both hospitals in our study had implemented simulation training for severe maternal morbidities, including eclampsia. This differs somewhat from a French study in which the lack of standardized protocols was discussed as one of the factors contributing to inadequate treatment with MgSO4. 18 In accordance with national guidelines for universal free pregnancy and postpartum care, follow‐up was also planned with the woman's general practitioner in 77% (17/22) of cases. 19 In line with previous care reviews from 1998 to 2000, the majority of eclampsia cases in our study occurred in a hospital setting. 11 Importantly, among all women with antepartum eclampsia occurring in the hospital, the diagnosis of preeclampsia was established before the seizure. It is worth considering whether some of these cases could have been prevented through different management. Many of the postpartum cases occurred in the labor and delivery unit, emphasizing the importance of close monitoring of blood pressure and neurological symptoms, especially in the early hours postpartum. Provider‐related delay was the most common factor contributing to delay in receiving appropriate care, including delay in recognizing and treating hypertension. 17 In addition, we noted that blood pressure was monitored within one hour prior to seizure in less than half of the hospital cases (8/18). Audit of maternal deaths due to hypertension in Norway from 1996 to 2014 pointed to failures of effective antihypertensive treatment and inadequate monitoring of blood pressure as the most common elements of substandard care. 20 Timely treatment of severe hypertension has been shown to reduce severe maternal morbidity and is also a marker for overall good care. 21 Similar to French and Dutch studies reporting low use of prophylactic MgSO4 in women with eclampsia (3%–7%), none of the women in our study received prophylactic MgSO4. 10 , 18 In line with our study, all the women in the Dutch study received MgSO4 treatment, whereas in the French study, this proportion was 82%. The national obstetric guideline recommends the use of MgSO4 prophylaxis for women with severe preeclampsia with rapid progression, also supporting clinical judgment in decisions to administer prophylaxis. 8 , 13 The MagPie‐trial showed a 50% reduction in eclampsia risk if women with severe preeclampsia were treated with prophylactic MgSO4. 9 As none of the women in our study received prophylactic MgSO4 it can be speculated that the use of MgSO4 is too restrictive in our setting. However, we had no data on the use of prophylactic MgSO4 in women who had severe preeclampsia/imminent eclampsia in our source population. The number needed to treat with MgSO4 to prevent one case of eclampsia ranges from 1:71 with severe preeclampsia to 1:36 when neurological symptoms are present. 22 An Irish study found a low eclampsia incidence with a restrictive use of prophylactic MgSO4, highlighting that the decision to initiate prophylaxis should be guided by the obstetricians' clinical judgment. 23 However, the working group concluded that prophylactic MgSO4 in five cases with established symptomatic preeclampsia could potentially have prevented eclampsia. Treatment with MgSO4 was delayed in two women, with the first seizure at home, and treatment was started in the hospital. In the plenary discussions, it was emphasized that MgSO4 should be available for administration in prehospital services. This may be particularly important in a country like Norway, where long travel distances to the obstetric unit are common. 24 Variation in the availability of MgSO4 within prehospital services is also seen in the United Kingdom, and has resulted in differences in the initial management of pregnant women presenting with seizures. 25 The incidence of eclampsia is declining in high‐income countries, 10 , 26 and current work in progress shows a similar decline in Norway; therefore, the likelihood of a non‐pregnancy‐related cause of seizure in pregnant and postpartum women is now higher than it was previously. Even though all women included in our study were discharged from the obstetric department with an ICD‐10 diagnosis of eclampsia (O15), five of them (23%) were found to have an alternative likely cause of the seizures. While it was not originally our aim to evaluate the correctness of the eclampsia diagnosis, this finding led to an important learning point: the necessity of thorough differential diagnostic evaluation of seizures in pregnant and postpartum women. Epilepsy was the most frequent alternative diagnosis in the non‐eclamptic group and represents both an important differential diagnosis and a potential long‐term complication associated with eclampsia, as an increased risk of epilepsy is reported after eclampsia. 27 , 28 In the three epilepsy cases, they were likely diagnosed with eclampsia as the seizure happened in relation to pregnancy and without a known diagnosis of epilepsy prior to the seizure. Another characteristic of the cases of epilepsy is that they all occurred intra‐ or postpartum, where triggers for an epileptic seizure, such as physical and mental stress and sleep deprivation, are present. 29 The audit identified differences in signs and symptoms between eclampsia and those with an alternative diagnosis. If a seizure occurs in relation to pregnancy and there are few typical signs of preeclampsia or eclampsia, it may very well be another cause than eclampsia. In addition, multiple seizures, particularly after treatment with MgSO4, should alert the clinician to do further diagnostic evaluation. 30 , 31 Importantly, after the initial management of a seizure, alternative causes of seizures must be considered. In our study population, 50% of the women had diagnostic imaging with cerebral CT or MRI after the seizure. The Norwegian guidelines for obstetric care do not address the need for cerebral imaging after eclampsia, though they recommend neurologic examination without specifying what this should include. 8 , 13 The general recommendation in cases of a new‐onset seizure is that all adults should have a cerebral CT or MRI as part of the diagnostic approach. 32 Our audit indicates that pregnant and postpartum women are less likely to receive cerebral imaging than non‐pregnant women. We raise the question whether or not cerebral imaging should be recommended regardless of a woman's pregnancy status in cases of new‐onset generalized seizures. Abnormal cerebral imaging is reported in 52%–63% of CTs and 73%–80% of MRIs after eclampsia. 33 , 34 , 35 The most common abnormal finding is posterior reversible encephalopathy syndrome, reported in 67% of women with eclampsia and characterized by vasogenic edema predominantly affecting the white matter substance of the posterior lobes. 36 Another study identified cerebral infarctions on MRI in 34% of women with eclampsia. 35 The lack of cerebral imaging in 50% in our study makes it challenging to evaluate differential diagnosis and neurological complications. Cerebral imaging and neurological follow‐up after suspected eclampsia have been recommended, as they are important for diagnostic clarification and may enable early initiation of specific treatment. 34 , 35 , 37 , 38 Postpartum follow‐up consultation with an obstetrician has been recommended after eclampsia in Norwegian obstetric guidelines since 2008. 8 , 13 , 14 In our study, we found that less than half of the women received or had planned a follow‐up appointment in obstetric specialist care. Women will likely seek explanations after experiencing a severe maternal morbidity. 39 A follow‐up consultation with an obstetrician may help debrief the event and reduce the risk of long‐term mental health illness. 40 Furthermore, it is important for future pregnancy planning, which is particularly relevant in our study population, where the majority (68%) of women were nulliparous. Also, among the women who were subsequently evaluated to have another more likely diagnosis, 3/5 had a follow‐up consultation with a medical specialty other than obstetrics. Due to the content of the case report form used for audit, it was not possible to determine the medical specialty for this follow‐up appointment. We received concise and uniform reporting of all cases using the digital CRF, which is a strength of our study. The extensive CRF comprised the most important aspects of the treatment course in women with eclampsia. The CRF was evaluated as user‐friendly by the collaborating clinicians, and it enabled the collection of clinical information in a time‐efficient manner. Our study population was small, and all cases were reported from two university hospitals in Western Norway. The obstetric units in these university hospitals are, however, among the four largest units in the country, and the clinical guidelines for hypertensive disorders in pregnancy are national; therefore, we believe the findings may be generalizable in Norway. The degree of generalizability to other countries depends on the similarity in guidelines and the organization of health care. While the CRF facilitates the systematic collection of similar information from all the cases, there might still be aspects we have missed in our evaluation. Our audit also revealed the need for further revision of the CRF where information on the treatment course was either missing or incomplete. Issues regarding documentation in the medical records, including challenges when reading hand‐written documentation, were also mentioned by the collaborating clinicians. As this is a retrospective audit, the evaluation of care may be influenced by hindsight bias. This was also part of the expert review discussions. In addition, the guidelines emphasize clinician discretion in initiating prophylaxis, which is challenging to evaluate retrospectively. Relying on discharge diagnosis of eclampsia may have resulted in missed cases of eclampsia due to miscoding or misdiagnosis and represents a potential source of underreporting. CONCLUSION From this audit of eclampsia, we were able to identify learning opportunities in 18 of 22 women. Improvements in obstetric care include a broad diagnostic approach to generalized seizures in pregnancy and postpartum, with routine cerebral imaging, as is the standard recommendation for new‐onset seizures in adults. Timely management of hypertension in pregnancy is essential to prevent delays in care. The current management practices of prophylactic MgSO4 in severe preeclampsia should be included in routine surveillance. Based on these findings, audit activities should continue and be integrated into the national routine surveillance of severe maternal morbidity. In future audits, patients' experiences and long‐term follow‐up should also be addressed. AUTHOR CONTRIBUTIONS Hilde Marie Engjom, Kari Klungsøyr, Siri Vangen, and Lill Trine Nyfløt contributed to the conception of this study. All authors participated in developing the variables used for data collection. The technical creation of the case report form was performed by Camilla Tjønneland Mentzoni. Data collection was performed by Kjerstine Røe, Marit Halonen Christiansen, Cecilie Fredvik Torkildsen, Tora Sandvik Fjeldstad, and Chen Sun. Data curation was done by Camilla Tjønneland Mentzoni. Truc Trang Nguyen and Hilde Marie Engjom developed the web‐based interface used for summarizing the information from the case report form. All authors contributed to reviewing the data for analysis and the interpretation of the results. Camilla Tjønneland Mentzoni wrote the original draft of the manuscript, and all authors reviewed and critically revised the manuscript. All authors approved the final version of the manuscript. FUNDING INFORMATION This study was funded by the Norwegian Research Council (NRC), research grant number 320181, the Norwegian SIDS and Stillbirth Society Grant/Award numbers 4/22 and 3/2019, and the Norwegian Medical Association's fund for quality improvement and patient safety, with grants for both the development and implementation phase. CONFLICT OF INTEREST STATEMENT All authors declare no conflict of interest. ETHICS STATEMENT The study was approved by the Regional Ethical Committee in the South‐East Health Region and was exempted from the principle of obtaining individual consent (reg. no. 248488, June 30, 2021). ACKNOWLEDGMENTS Norwegian Women's Public Health Association participated in the eclampsia working group with a patient and public perspective. OCINOR—Eclampsia Working Group: Victoria Bakke (Department of Anesthesiology and Intensive Care, Hospital of Vestfold, Tønsberg, Norway), Tone Bolstad (Department of Anesthesiology and Intensive Care Medicine, Haukeland University Hospital, Bergen, Norway), Kristin Fagereng (Department of Prehospital Care, Stavanger University Hospital, Stavanger, Norway), Inger Johanne Ravn (Norwegian Association of General Practice), Ragnar Kvie Sande (Department of Obstetrics and Gynecology, Stavanger University Hospital, Stavanger, Norway; Department of Clinical Science, University of Bergen, Bergen, Norway), Vibeke Stuestøl (Norwegian Association of General Practice), Karoline Sunnarvik‐Ween (Department of Gynecology and Obstetrics, Drammen Hospital, Drammen, Norway), Ingvil Krarup Sørbye (Institute of Clinical Medicine, University of Oslo, Oslo, Norway; Division of Obstetrics and Gynaecology; Department of Obstetrics, Oslo University Hospital, Oslo, Norway), Cecilie Fredvik Torkildsen (Department of Obstetrics and Gynecology, Stavanger University Hospital, Stavanger, Norway; Department of Gynecology and Obstetrics, Drammen Hospital, Drammen, Norway). Contributor Information Camilla Tjønneland Mentzoni, Email: camillatjonneland.mentzoni@fhi.no. OCINOR—Eclampsia Working Group: Victoria Bakke, Tone Bolstad, Kristin Fagereng, Inger Johanne Ravn, Ragnar Kvie Sande, Vibeke Stuestøl, Karoline Sunnarvik‐Ween, Ingvil Krarup Sørbye, and Cecilie Fredvik Torkildsen DATA AVAILABILITY STATEMENT The data cannot be shared due to confidentiality regulations. REFERENCES Associated Data Data Availability Statement The data cannot be shared due to confidentiality regulations.