Development of obstetric-hepatology services: defining optimal care and impact of pregnancy counselling on patient experience

Abstract

Objective

Prepregnancy counselling (PPC) is an important aspect of care for women with chronic liver disease (CLD) and liver transplantation (LT), yet its impact has not been well described. This study aims to assess the experience of women attending a joint obstetric-hepatology PPC clinic in a single-centre unit.

Design/methods

A retrospective questionnaire-based study in a tertiary unit within the UK where patients who attended the PPC clinic between March 2016 and July 2021 were invited to participate by filling in a questionnaire. Descriptive data and free-text content were subsequently analysed.

Results

108 women attended the PPC clinic over a 5-year period. Overall, 58/108 (54%) completed the questionnaire. Principal concerns regarding pregnancy included fears around deterioration in health (66%), maternal death (24%), pregnancy loss (66%), medication effects (60%) and disease transmission (36%). 17/58 (14%) patients felt the presence of multiple doctors was intimidating, however, perceptions improved by the end of the consultation.

Overall, 44/58 (76%) respondents felt the clinic helped them reach a decision about pursuing pregnancy. Almost all respondents would recommend the clinic to others. There were no major differences in pregnancy outcomes between those that received PPC and those that did not.

Conclusion

The PPC clinic facilitates a personalised approach to care and is well received by patients with CLD/LT. It is difficult to elucidate whether attendance alone impacts on pregnancy outcomes; registry data may be better placed at addressing this important question.

What is already known on this topic

• The impact of prepregnancy counselling (PPC) in hepatology patients is not well defined.

What this study adds

• This questionnaire-based evaluation of PPC in patients with chronic liver disease (CLD)/liver transplantation (LT) has demonstrated high levels of patient satisfaction.

How this study might affect research, practice or policy

• The impact of PPC on pregnancy related outcomes remains to be addressed.

• PPC should be embedded into clinical practice for women of reproductive age with CLD/LT.

Introduction

The number of women with chronic liver disease (CLD) or liver transplantation (LT) achieving pregnancy has risen.^1–6 CLD includes conditions which cause progressive destruction of the liver over time leading to fibrosis and cirrhosis. Hepatic decompensation can complicate pregnancy in those with advanced CLD.⁷ In LT recipients, pregnancy may worsen renal function and increase the risk of pre-eclampsia.^{8 9} Fetal complications in pregnant women with CLD and/or LT include preterm delivery, low birth weight, intrauterine growth restriction and neonatal distress syndrome.^10–12 As such, pregnancies in women with CLD/LT require specific considerations and prepregnancy counselling (PPC) provides an opportunity to address these.

Women with pre-existing medical conditions should receive PPC by doctors experienced in managing these disorders during pregnancy.^{13 14} Guidelines in autoimmune hepatitis (AIH) recommend that the effects of disease and medication on materno-fetal health should be discussed prior to pregnancy.¹⁵ By doing this, conception can be planned at a time of stable disease activity while on appropriate medication. Stable liver disease at conception is more common in women who receive PPC, although PPC may not specifically affect live birth or prematurity rates.¹⁶

The severity of CLD and the risk of potential complications in future pregnancies can be assessed. Women with cirrhosis who receive PPC are more likely to have endoscopic variceal screening.¹⁶ This facilitates informed decision making and a personalised approach to care with timely interventions in selected patients.

PPC and its impact on women with CLD/LT, including pregnancy outcomes, are not well described in hepatology.⁹ The aim of this study was to evaluate our experience of women with CLD/LT who received PPC in a single-centre unit and identify areas for improvement.

Methods

The PPC clinic for women with CLD/LT was established in our unit in March 2016. Referrals are received from clinicians across the South of England. The appointment typically lasts 30 min, during which women are reviewed by an obstetrician, obstetric-physician and hepatologist who provide individualised and evidence-based care.

Pregnancy outcomes and effects on hepatic/graft function are discussed with the patient. Medications are reviewed for teratogenicity. Prophylactic aspirin is recommended for patients at high risk of pre-eclampsia.¹⁷ In patients with portal hypertension, endoscopic variceal screening is arranged for the mid-second trimester. Local or tertiary antenatal care and delivery are planned. A comprehensive letter is sent to the patient, referrer and general practitioner reiterating the advice given.

The experience of women who attended this clinic was evaluated using a questionnaire adapted from a study in PPC and chronic kidney disease.⁹ This questionnaire was based on a framework from the Institute of Medicine focussing on patient-centred care.¹⁸ The final questionnaire included 28 fixed closed-format questions examining women’s views on their involvement, shared decision-making and communication of information. A five-point response scale was used to maximise variability.¹⁹ Free-text responses were invited for recommendations to improve their experience. The questionnaire was anonymised to encourage response and honesty (online supplemental material).

Between February 2019 and July 2021, the questionnaire was posted with a paid return envelope to all clinic attendees between March 2016 and July 2021. Questionnaires were sent with an explanation of the study, voluntary nature of participation and anonymity of response (online supplemental material). Follow-up telephone calls were made to remind patients to participate. Though the letter did not specify the implications of not completing the questionnaire, consent was implied with the return of a completed questionnaire.

For clinic attendees who subsequently delivered at our hospital, the electronic patient records were retrospectively analysed for pregnancy outcomes including gestational diabetes/hypertension, pre-eclampsia, fetal growth restriction (<10th centile for gestational age), preterm delivery (<37 weeks gestation) and intrauterine death (IUD, ≥24 weeks gestation).

A control group of 75 women with CLD/LT who had undergone pregnancy between 2016 and 2021 but not attended PPC clinic was derived from our database. The control group was matched for age and those with cirrhosis had comparable Model for End-stage Liver Disease (MELD) scores. Statistical analysis was performed with Fisher’s exact test and p values <0.05 were deemed statistically significant.

Results

Between March 2016 and July 2021, all 116 patients referred (via email or written correspondence from obstetric, gastroenterology or hepatology teams) to the 3 clinicians running the PPC were invited to attend clinic. In the study period, 108/116 women attended clinic and 58/108 patients (54%) responded to the questionnaire. Seventy-five women attended clinic once and 33 women (28%) attended more than once based on clinical need±patient choice (range 1–7), equating to a total of 162 appointments over 5 years. Demand for appointments steadily increased until the COVID-19 pandemic (online supplemental material). There were no significant differences in demographic data or clinical outcomes between those that responded to the questionnaire and those that did not.

Demographics

The median age of women offered an appointment was 32 years (range 24–50). Ethnicities (n=116) included: white British (53%), Black British/African/Caribbean (17%), British Indian/Pakistani (11%), white European (9%), Middle Eastern (6%), South American (2%) and other (3%).

Common aetiologies in non-LT recipients included AIH (24%, 21/86) and primary sclerosing cholangitis (13%, 11/86). Ten women had cirrhosis. The most common pregnancy-specific condition was intrahepatic cholestasis of pregnancy (ICP) (21%, 18/86). Twenty-six per cent (30/116) of cases were LT recipients (online supplemental material).

Questionnaire data

Before the appointment

Prior to their appointment, 64% (37/58) had previously discussed impact or implications of pregnancy with their general practitioner or hepatologist. Only 47% (27/58) felt well informed about how their liver condition might affect a future pregnancy (figure 1A). Despite 74% (43/58) of women hoping for a healthy pregnancy, only 34% (20/58) felt confident about achieving one (figure 1B). Concerns regarding future pregnancy included medication effects on the fetus (60%), pregnancy loss (66%) and deterioration in maternal health (66%) (figure 2). These concerns prevented 17/58 (29%) women from pursuing a pregnancy prior to clinic.

Figure 1.

Responses regarding perceptions prior to clinic, structure of consultation and impact on decision-making.

Figure 2.

Patient’s fears and concerns regarding a potential future pregnancy. *Others=concerns about fertility and ability to conceive, effects of pregnancy on mental health, recurrence of previous pregnancy-related complications, effects of early delivery on baby, having to require caesarean section delivery when vaginal delivery is preferred option and impact of fibroids on delivery.

During the appointment

By the time of the consultation, 48% (28/58) of women were already pregnant (first trimester=14/28; second trimester=8/28; third trimester=4/28; post partum=2/28). The remaining women received prepregnancy advice. Most women (71%, 41/58) were not intimidated by the number of clinicians in the room and found the number of experts reassuring (93%, 54/58) (figure 1C, D).

Almost all patients (95%) felt that their concerns were listened to (figure 1E). Fertility issues and assisted conception were discussed in 9 cases (16%).

All women taking regular medication (n=47) were informed about their safety during pregnancy. Altogether, 40% (19/47) had their medications altered including a change from mycophenolate mofetil to tacrolimus in one transplant recipient, three patients (one AIH, two LT) had immunosuppression escalated, eight women had their ursodeoxycholic acid dose increased, two patients had rifampicin added and three patients had their antihypertensives or anticoagulant changed. Aspirin was initiated in two women as pre-eclampsia prophylaxis. Safety of continuing medication during breastfeeding was discussed in 19% (9/47) of women.

After the appointment

Overall, 57/58 women found the clinic informative. All patients understood the information provided (figure 1F). Most patients considered the information given reassuring (71%, 41/58) while 14% (8/58) found the information worrying (figure 2). Of those who found the information worrying, 3 had cirrhosis with portal hypertension (MELD 9–12). Of those who found the information reassuring, 5 had portal hypertension (cirrhosis: 4 (MELD 7–10), non-cirrhotic: 1).

The information given in clinic helped 76% (44/58) of women to decide on whether to embark or continue pregnancy (figure 1G). Furthermore, 83% (48/58) decided to attempt conception or continue with pregnancy if already pregnant.

Finally, 14% (8/58) of patients reported not receiving a letter after clinic (figure 3C). Those who received a letter found it useful. By the end of the consultation, one patient still felt intimidated by the number of specialists in clinic. Ninety-five per cent (55/58) of women felt that they had received the best possible care (figure 3B). All but one patient would recommend the clinic to others.

Figure 3.

Responses to questions regarding postclinic experience.

Free-text responses

Half the respondents provided feedback in the free text box with mostly positive comments: ‘level of thought and care was very comforting…the clinic exceeded my expectations’ and ‘after the terrible experience I had with my first baby (nearly losing my life and having a LT), the team made the whole experience more positive…they were vigilant and monitored me closely’. One patient ‘wished’ they had been reviewed sooner because the clinic provided answers to questions and reassurance.

One patient, who experienced poor pregnancy outcomes, felt that the doctors should not ‘rely heavily on the medical model’ but focus on the patient’s previous experiences and respect their wishes to continue pregnancy.

Two women suggested the appointment letter could explain the purpose of the appointment, who would be present in the consultation and whether the patient’s partner could attend.

Five women felt the service could be improved by providing a follow-up appointment or contact details if further questions needed to be addressed.

Pregnancy outcomes

Data on pregnancy outcomes were missing in 34/108 patients. In those who achieved pregnancy beyond 24 weeks gestation (n=56), rates of gestational diabetes and hypertension were 5%. Pre-eclampsia rates were 9% in the group that attended clinic (control group 3%, p=0.24). ICP rates were greater in the group that attended clinic compared with the control group (30% vs 6%, p=0.0005). Hepatic decompensation and postpartum flares of AIH (2 vs 4, p=0.69) were similar in both groups. Admission rates and length of stay in the intensive care unit (ICU) were similar between the two groups (clinic attendees: median 3 days, range 1–120 days vs control group: median 3.5 days, range 1–109 days). There were no maternal deaths during pregnancy. Materno-fetal complications are summarised in table 1.

Table 1.

Maternal and fetal outcomes in women who have attended a pregnancy counselling clinic and those who have not

	Attended clinic (n=74)	Control group (n=75)	P value
Demographics
Median age (range) in years	31 (24–50)	30 (22–47)	–
Ethnicity	White British 61% Black British/African/Caribbean 22% Other 17%	White British 57% Black British/African/Caribbean 20% Other 23%	–
Severity of disease in those with cirrhosis (median MELD score and range)	9 (6–20) (n=10)	8 (6–18) (n=10)	–
Pregnancy outcomes
No pregnancy	18% (13/74)		–
Achieved pregnancy	82% (61/74)		–
Miscarriage	4 (early)	7 (early)	NS (0.75)
Intrauterine death (IUD)	2 IUDs at 23+4 and 32 weeks	1 IUD at 34 weeks	NS
Maternal complications in pregnancies ≥24 weeks
Gestational diabetes	5% (3/56)	7% (5/68)	NS (0.73)
Gestational hypertension	5% (3/56)	4% (3/68)	NS (0.99)
Pre-eclampsia	9% (5/56)	3% (2/68)	NS (0.24)
Intrahepatic cholestasis of pregnancy	30% (17/56) (onset 14–35 weeks)	6% (4/68) (onset 24–37 weeks)	0.0005
Postpartum haemorrhage	29% (16/56)	12% (8/68)	0.02
Hepatic decompensation	Two variceal bleeds during pregnancy One postdelivery hepatic decompensation with ascites and variceal bleeding) Two post-partum flares of autoimmune hepatitis	One variceal bleed during pregnancy One hepatic decompensation with ascites in the context of in vitro fertilisation One variceal bleed post-partum Four postpartum flares of autoimmune hepatitis	NS for decompensation (0.99) NS for postpartum flare (0.69)
Maternal intensive care	7% (4/56)	4% (3/68)	NS (0.70)
Fetal complications
Live birth	90% (55/61)	89% (67/75)	NS (0.99)
Median gestation duration (weeks)	37+6 (32 to 42+2)	38+2 (33+2 to 41+3)	–
Term (≥37 weeks gestation)	67% (37/55)	72% (48/67)	NS (0.69)
Prematurity	33% (18/55)	28% (19/67)	NS (0.69)
Median birth weight (grams)	3056 (range 1724–4375)	3321 (range 2545–3675)	–
Median birthweight centile	44 (range 1–100)	48 (range 12–85)	–
Neonatal intensive care or special care baby unit	22% (12/55) one neonatal death	6% (4/67) No deaths	0.01

Unable to assess as patients were identified through an earlier prospectively collated database, i.e. not seen in pre-conception phase.

Bold values denote significant values.

MELD, Model for End-stage Liver Disease; NS, non-significant.

Of the live births (n=55), vaginal deliveries occurred in 51% (28/55) of women who attended clinic (control group 64%). Elective caesarean sections occurred in 29% (16/55) of those who attended clinic (control group 24%).

Live birth rates (≈90%, p=0.70) and median gestational length (≥37 weeks) were similar between the group that attended clinic vs those that did not. The proportion of preterm births were similar (28%–33%) in both cohorts. Median birth weight in the group that attended clinic was 3056 gs (range 1724–4375) and 3321 gs (range 2545–3675) in the control group.

Admission rates to neonatal ICU were higher in the group that attended clinic compared with the control group (22% vs 6%, p=0.01). Of the 12 neonates requiring admission to high-dependency unit (HDU), 8 were preterm (32 to 36+3 weeks gestation). Seven cases only required a brief stay. One preterm baby with suspected hydrops fetalis required multiple intrauterine transfusions. After birth, he stayed in the neonatal ICU for 15 days, HDU for 19 days and special care baby unit for 67 days. He survived and was discharged.

There was one neonatal death in a mother who had undergone in vitro fertilisation and was known to have a large uterine mass (probable fibroid) prior to pregnancy. At 17 weeks gestation, a chorioangioma was identified. At 22+6 weeks gestation, the woman experienced premature rupture of membranes. She went into preterm labour at 32+5 weeks gestation. The baby had group B streptococcus septicaemia and died after several weeks.

A summary table of pregnancy outcomes between women with cirrhosis who attended PPC clinic versus the control group who had not attended can be found in online supplemental material. Statistical analysis was not performed due to the small number of patients in each group.

Discussion

This questionnaire-based evaluation of patient experience of the PPC clinic has demonstrated a high level of patient satisfaction. The majority found the clinic informative, and for some it influenced their decision-making process. The multidisciplinary presence of specialists was a positive experience for 93% of respondents. Only one patient felt intimidated by the number of clinicians. This could be improved by mentioning who will be present in clinic on the appointment letter.

PPC was delivered in a joint obstetric-hepatology clinic that requires each specialist to be available on the same day and the clinic to be conducted in an appropriate space. This may not be feasible in other centres. In such cases, close teamwork between hepatologists and obstetricians to deliver PPC to patients is key.

Two patients gave very positive responses to the questions, including the free text box, and their single negative response was incongruous with the rest of their responses. This may have been purposeful, or alternatively the 5-point response scale resulted in confusion. One patient gave consistently negative responses. This woman was on the LT waiting list with an MELD score of 20, large varices (including parametrial) and complex cardiac issues. At the first clinic appointment, she was already pregnant, and the high risk of decompensation was explained. She was keen to continue the pregnancy due to a previous successful pregnancy(when her liver disease was less severe. At 32 weeks, she experienced severe per vaginal bleeding and had a cardiac arrest. She underwent emergency abdominal surgery and lost the pregnancy. The questionnaire was completed shortly after discharge. It is possible that poor pregnancy outcomes and difficulties conceiving may influence questionnaire responses.

Furthermore, questionnaire respondents may be more proactive individuals invested in their own health, thus introducing bias into answers. Although not evaluated, potential reasons for not returning the questionnaire include outdated contact details, language barriers and patient choice. In the future, the questionnaire can be developed in multiple languages and the cover letter can specify that completion is voluntary and will not affect patient care.

Pregnancies in the PPC group had similarly high rates of materno-fetal complications compared with the control group. This may partly be explained by 48% of women already being pregnant at their first PPC appointment, thus missing the opportunity to modify risk factors prior to conception. The small size of the study from a single-centre unit also makes drawing comparisons difficult. The high rates of ICP cases in clinic may be secondary to the presence of an obstetric-physician with a specialist interest in the disease, thus precipitating numerous referrals. Rates of gestational diabetes/hypertension, pre-eclampsia, hepatic decompensation, postpartum haemorrhage, live births and prematurity are comparable to those reported in the literature.^{10 20} However, there was a high rate of babies requiring HDU care. One explanation is that patients with complicated liver disease may be more likely to attend clinic than those with straightforward disease, thus resulting in a ‘sicker’ cohort of patients attending clinic. There was an attempt to address this in the control group by matching age and MELD scores. However, complicated liver disease may not be wholly reflected by MELD scores and other factors such as the presence of comorbidities, medication use and non-adherence to advice need to be considered. Furthermore, pregnancy outcomes were missing in some women because they delivered in a different hospital. There is need for further data on the impact of PPC on pregnancy outcomes.

The clinic can be better used with earlier referrals, so that patients can be optimised prior to conception. Benefits of PPC with data on pregnancy outcomes would help increase awareness among clinicians to proactively ask patients of childbearing age regarding family planning so that early referrals can be made. More attention should be given to contraception, assisted conception and breast feeding, for example, medication use during breast feeding was only discussed in 19% of women in this study. Contributing factors may include the lack of midwife input during the consultation and an assumption that this would be discussed later once pregnancy had been achieved. Postclinic, 14% of patients did not receive a letter, despite evidence of a typed letter on electronic records. A follow-up telephone call or email address may be useful for those who have further questions after clinic, but this would require appropriate resource management.

Conclusion

This study demonstrated high levels of patient satisfaction in women with CLD/LT who attended the PPC clinic. Women were given information on individualised risk profiles and what to expect from pregnancy which subsequently allowed informed decision-making. However, it remains to be seen whether PPC itself influences pregnancy outcomes.

Data availability statement

All data relevant to the study are included in the article or uploaded as online supplemental information. The data that support the findings of this study are available from the corresponding author, MAH, upon reasonable request.

Ethics statements

Patient consent for publication

Consent obtained directly from patient(s)

Ethics approval

This was a service evaluation project which did not require formal NHS Research Ethics Committee review. There was no deviation from normal clinical practice for these patients and participants were informed of the study objectives. Our study was registered and approved as an audit at King’s College Hospital.