Meta-analysis: the impact of disease activity at conception on disease activity during pregnancy in patients with inflammatory bowel disease

Background

The rate of IBD exacerbation during pregnancy varies in the published literature.

Aim

We sought to perform a systematic review and meta-analysis of the effects of disease activity at conception on disease course during pregnancy in women with IBD.

Methods

Published studies and abstracts from standard sources were screened for appropriate studies. Data was pooled and analyzed using funnel and forest plots. Quality assessment scores were given using GRADE criteria.

Results

Fourteen studies were eligible for inclusion; ten studies contained patients with UC (N=1130), and six studies contained patients with CD (N=590). In patients with UC there was a significantly higher risk ratio of active disease during pregnancy in patients who commenced pregnancy with active disease (55%), when compared with those in remission at conception (36%) (RR 2.0, 95% CI: 1.5–3, p<0.001). This risk was also higher in patients with CD, (RR 2.0, 95% CI: 1.2–3.4, p=0.006). Thirteen of the studies rated “low” in all domains of a quality assessment, and there was significant statistical heterogeneity.

Conclusions

Patients with IBD who conceive when their disease is active are more likely to have active disease during pregnancy than those who conceive when in remission. All studies used in this analysis had a high risk of bias therefore further studies are required.

INTRODUCTION

Inflammatory Bowel Disease (IBD) is a chronic condition that is often diagnosed in the first half of life; fifty percent of patients with IBD are diagnosed <35Y(1). This period coincides with the peak fertility years for young women planning pregnancy. This population of patients is naturally interested in the possible effects of a pregnancy on the activity of their IBD when family planning (2).

Initial reports dating back to the 1950s painted a grim picture, with up to 90% of patients relapsing during pregnancy (3). Subsequent studies since the 1980s have described relapse rates of 30–50% during pregnancy, which were still higher than the rate in non-pregnant women in some reports (4,5). This issue has clearly influenced the pregnancy plans of some women with IBD over the last 50 years. A study of women with IBD found that 46% of women reported that the disease had “changed their attitudes toward child-bearing”, and 16% were voluntarily childless, compared to 6.2% in the general population (6). Most of the voluntarily childless women in this study responded that the concern that their IBD would worsen during pregnancy would “discourage them from having children”. Women diagnosed with IBD also tended to have fewer subsequent pregnancies (mean 1.7) than women who had their first pregnancy before IBD was diagnosed (mean 2.4) (7).

In contrast, more recent large studies have suggested that there is no significant change in disease severity during pregnancy, when the pre- and post- conception periods are compared (8). Similarly, Nielsen et al. reported that the overall risk of disease exacerbation for women was 32% per year during women’s fertile years, and 34% per year during pregnancy (5). However, it is clear from the historical literature that disease activity at conception was an important predictor of disease outcome during pregnancy (9,10). If IBD is inactive at conception, the chance of disease flare was equal to that of non-pregnant patients (11). However, if IBD was active at conception there was 60% chance of the disease remaining active or worsening during the pregnancy (10). Expert consensus now emphasizes disease control prior to conception for these reasons (12).

It is clear that the management and natural history of IBD has changed since the 1950s, so the concerns about pregnancy’s effects on disease activity may be over-estimated (13). In light of this, and of the wide range of effect estimates of pregnancy on disease activity in the literature, we sought to perform a systematic review and meta-analysis of the effect on disease activity at conception on disease course during pregnancy in women with IBD.

METHODS

Literature Search

A literature search was performed to identify all published and unpublished studies in any language that reported disease activity at conception and during pregnancy in patients with IBD. A systematic search of the following databases was performed: MEDLINE (Pubmed) – 1966 to January 2013, Web of Science - 2000 to January 2013), Allied Health Literature (CINAHL) – 1990 to January 2013, Scopus – 2000 to January 2013, and EMBASE 2000 – January 2013. The following search strategy was constructed by using a combination of MeSH subject headings and text-words; “Inflammatory Bowel Disease”, “Crohn’s disease,” “Ulcerative colitis,” “pregnancy,” “conception,” and “disease activity.” Digital records of abstracts from American Digestive Diseases Week and the United European Gastroenterology Week (2006–2012) were also searched using these search terms, and reference lists of all articles read and several previously published reviews were scrutinized to disclose additional literature on the topic.

Inclusion and Exclusion Criteria

We included all studies (controlled trials, observational studies, cohort studies) that reported disease relapse rates or clinical outcomes during pregnancy according to disease activity at conception in women with IBD. The primary outcome measure was the proportion of patients with “active disease” during the 9 months of pregnancy, grouped according to level of disease activity at conception (active or inactive). No pre-specified clinical disease activity score was used to determine this outcome, as we expected there would be no universal definition of “active disease” across the studies beyond physicians’ global assessment. We did not require that included studies report the objective confirmation of active inflammation as the cause of symptoms. Studies that measured “active disease” by any method were included if they also reported these outcomes according to the patients’ disease activity at conception.

We excluded studies if they a) were review articles b) did not report disease activity during pregnancy, c) did not report disease activity at conception / baseline, d) pooled outcomes from the post-partum period with the pregnancy period.

Study Selection

Two authors (AA and MH) independently scanned the abstract of every trial identified by the search to determine eligibility. Blinding to source was not performed. Full articles were selected for further assessment if the abstract suggested the study included patients with IBD, who were followed during pregnancy, and both disease activity at baseline, and during pregnancy were recorded and reported. If these criteria were unclear from the abstract, the full article was retrieved for clarification. Papers not meeting the inclusion criteria were excluded. Any disagreements were resolved by discussion, and if required, by consultation with the senior author (ACM).

Data Extraction and Quality Assessment

The following data were retrieved (where possible) from published reports using standardized forms with disagreements resolved by discussion between the reviewers: number of patients in the study, method of selection of patients, numbers of patients with active and inactive disease at conception, methods of measurement of clinical outcomes, and reporting outcomes used. For observational studies, methodological quality was assessed by determining the eligibility criteria, degree of measurement of exposure and outcome, control for confounders, and completeness of follow up (14).

Statistical Analysis

Data were analyzed and reported consistent with the consensus guidelines by the Meta-analysis of Observational Studies in Epidemiology group (15). Data were pooled for meta-analysis if the outcomes were sufficiently similar (determined by consensus of authors) and data were homogenous (determined by the degree of clinical and statistical heterogeneity). Data were pooled for meta-analysis if the criteria for “active disease” in each study were determined by consensus to represent clinically comparable events in the context of disease activity. Raw data from included studies (absolute numbers) were used to construct 2 × 2 contingency tables, and unadjusted risk ratios (RRs) were calculated using Review Manager (RevMan 5.1) for dichotomous outcomes. The random effects model was used to account for variations between studies and give a more conservative pooled estimate (16). The Q test was used to assess for heterogeneity and I ^{statistic to quantify the percentage of heterogeneity due to between-study variation; a value of P < 0.10 was considered statistically significant (}17). Funnel plots and the Egger's test were used to evaluate for publication bias (18). The number needed to harm (NNH) was calculated where informative (19). Sensitivity analyses were performed for all outcomes where 3 or more studies were included. Sensitivity analysis included sensitivity based on individual studies, sample size, method of disease activity assessment, specialist or community care. Significance levels were set at a P < 0.05.

RESULTS

Study Characteristics

Figure 1 illustrates the study selection process; of the 26 references identified from database searches, 25 articles were selected for full text review. From these, only 14 articles were deemed suitable and met the pre-specified inclusion criteria. Twelve articles/abstracts were excluded for various reasons detailed in supplementary table 1. Table 1 summarizes the main characteristics of the 14 included studies (3,5,9,11,20–29). Patients with Crohn’s disease (CD) were included in 6 studies (N= 590), and patients with Ulcerative Colitis (UC) were included in 10 studies (N=1130). Length of follow up in these studies ranged from 1 to 12 years. The primary outcome for inclusion in this meta-analysis, “active disease”, was reported based on clinicians’ assessment or Harvey Bradshaw Index (HBI) in patients with Crohn’s disease, and clinicians’ assessment, Truelove criteria or Simple Colitis Clinical Activity Index (SCCAI) in patients with UC. In the quality assessment of included studies, all 14 studies were graded for quality (supplementary table 2).

Figure 1

Flow-chart of filtering of search results

TABLE 1

Characteristics of included studies

Study ID	Year	N	Follow-up	Disease Activity Measure
Abramson	1951	37	6 yrs	Physician assessment
Banks	1957	69	12 yrs	Physician assessment
Bortoli	2011	332	4 yrs	SCCAI (UC) or HBI (CD)
Crohns	1956	112	12 yrs	Physician assessment
DeDombal	1965	83	1 yr	Truelove criteria
Homan	1976	43	Age 45	N/A
Khosla	1984	72	< 1 yr	N/A
Macdougall	1956	81	< 1 yr	N/A
McEwen	1972	47	< 1 yr	Truelove criteria
Mogadam	1981	324	< 1 yr	N/A
Morales	2000	35	< 1 yr	Physician assessment
Nielson	1983	152	15mths	Physician assessment
Willoughby	1980	184	< 1 yr	N/A
Woolfson	1990	78	< 1 yr	N/A

Effects of Disease Activity at Conception on Disease Activity during Pregnancy

Ten studies reported raw data on patients with UC for meta-analysis (3,5,11,20–23,25,26,28). Amongst the 1130 patients with UC, 55% of patients who became pregnant when their UC was active remained active during the pregnancy, as compared to a rate of relapse of 29% amongst women who became pregnant in remission. The risk ratio for active disease during pregnancy was 2.0 in this cohort in the meta-analysis (95% CI: 1.5–3, p<0.001, Figure 2), when compared to patients in remission at conception. The test for overall effect was significant, but there was significant statistical heterogeneity, with an I ^{of 64%.}

Figure 2

Forest plot of meta-analysis of “active disease during pregnancy” according to patients’ disease activity at conception in those with Ulcerative Colitis. The squares and lines represent the effect estimate and confidence intervals for each individual study, and the diamond represents the overall effect estimate for the pooled studies. If the effect estimate is to the right of the vertical line in the forest plot, it means that the “active” group had a higher RR of disease activity (the negative outcome) during pregnancy, so this favors being “inactive” at conception.

Six studies of patients with CD reported raw data for meta-analysis, including 519 patients(9,11,21,24,27,29). In the studies with patients with CD, 46% of patients who had active disease at the time of conception remained active, compared with relapses in 23% of patients who got pregnant while in remission. The risk ratio for active disease was 2.0 (95% CI: 1.2–3.4, p=0.006, Figure 3). There was significant statistical heterogeneity, with an I ^{of 59%.}

Figure 3

Forest plot of meta-analysis of “active disease during pregnancy” according to patients’ disease activity at conception in those with Crohn’s disease.

Study Characteristics

Figure 1 illustrates the study selection process; of the 26 references identified from database searches, 25 articles were selected for full text review. From these, only 14 articles were deemed suitable and met the pre-specified inclusion criteria. Twelve articles/abstracts were excluded for various reasons detailed in supplementary table 1. Table 1 summarizes the main characteristics of the 14 included studies (3,5,9,11,20–29). Patients with Crohn’s disease (CD) were included in 6 studies (N= 590), and patients with Ulcerative Colitis (UC) were included in 10 studies (N=1130). Length of follow up in these studies ranged from 1 to 12 years. The primary outcome for inclusion in this meta-analysis, “active disease”, was reported based on clinicians’ assessment or Harvey Bradshaw Index (HBI) in patients with Crohn’s disease, and clinicians’ assessment, Truelove criteria or Simple Colitis Clinical Activity Index (SCCAI) in patients with UC. In the quality assessment of included studies, all 14 studies were graded for quality (supplementary table 2).

Figure 1

Flow-chart of filtering of search results

Effects of Disease Activity at Conception on Disease Activity during Pregnancy

Ten studies reported raw data on patients with UC for meta-analysis (3,5,11,20–23,25,26,28). Amongst the 1130 patients with UC, 55% of patients who became pregnant when their UC was active remained active during the pregnancy, as compared to a rate of relapse of 29% amongst women who became pregnant in remission. The risk ratio for active disease during pregnancy was 2.0 in this cohort in the meta-analysis (95% CI: 1.5–3, p<0.001, Figure 2), when compared to patients in remission at conception. The test for overall effect was significant, but there was significant statistical heterogeneity, with an I ^{of 64%.}

Figure 2

Forest plot of meta-analysis of “active disease during pregnancy” according to patients’ disease activity at conception in those with Ulcerative Colitis. The squares and lines represent the effect estimate and confidence intervals for each individual study, and the diamond represents the overall effect estimate for the pooled studies. If the effect estimate is to the right of the vertical line in the forest plot, it means that the “active” group had a higher RR of disease activity (the negative outcome) during pregnancy, so this favors being “inactive” at conception.

Six studies of patients with CD reported raw data for meta-analysis, including 519 patients(9,11,21,24,27,29). In the studies with patients with CD, 46% of patients who had active disease at the time of conception remained active, compared with relapses in 23% of patients who got pregnant while in remission. The risk ratio for active disease was 2.0 (95% CI: 1.2–3.4, p=0.006, Figure 3). There was significant statistical heterogeneity, with an I ^{of 59%.}

Figure 3

Forest plot of meta-analysis of “active disease during pregnancy” according to patients’ disease activity at conception in those with Crohn’s disease.

Sensitivity Analysis

Sensitivity analysis was performed for the overall effect estimate for outcomes where there were > 3 studies in a meta-analysis. Analyses based on effect estimates (fixed or random), year of publication (pre-1980 or post 1980), criteria for assessment (physicians’ assessment or SCCAI/HBI) did not significantly change the effect estimate. No adjusted data was reported in the included studies for further subgroup analyses.

In order to identify the determinants of the statistical heterogeneity seen in the main sub-group analysis, each individual study was excluded, and the I ^{value measured for heterogeneity. The main source of statistical heterogeneity in the meta-analysis of the UC studies was the paper by Crohn’s et al., whereas the main source in the CD studies was the study by Bortoli et al. There was no obvious clinical heterogeneity between these studies and the other included studies based on patients’ populations or outcomes.}

Tests for Publication Bias

Formal tests for funnel plot asymmetry were not performed as the power of these tests is too low to distinguish chance from real asymmetry if less than 10 studies are included.

DISCUSSION

Women are understandably often anxious about how their IBD may behave during pregnancy. In addition, they often also have concerns about continuing their IBD medications during pregnancy, due to potential effects on the fetus (30). As a consequence, self-reported levels of medication adherence during pregnancy are only modest, primarily due to this fear of detrimental consequences for the fetus (30,31). Such non-adherence has been associated with a higher risk of disease exacerbation in the peri-partum period (32). Gastroenterologists have historically encouraged the use of maintenance therapy when pregnancy is considered, as prior studies have emphasized the risk of persistent disease activity in patients who become pregnant when their IBD is active (1,11,33).

This is the first meta-analysis to determine the effect estimate of disease activity at conception on the course of IBD during pregnancy. In patients with IBD, there is a 2-fold risk of colitis remaining persistently active during pregnancy if patients become pregnant with active colitis. Surprisingly, the effect estimate was not significantly different when different decades were compared; suggesting that addressing active disease during pregnancy still remains a challenge. There was a

These results are consistent with a recent large study on this topic from a European multi-center cohort; 14% of those in remission at conception relapsed during pregnancy, whereas 26% of those active remained so until delivery (21). These rates were lower than the largest prior study, published in 1981; 25% of those in remission relapsed during pregnancy, whereas 51% of those active remained so until delivery (11). Many of the early reports on this outcome came from referral centers, and there has been a deficit of population-based studies in less severe cases of IBD.

One note of interest from these studies is that about half of the patients who became pregnant when their IBD was active did not remain active during the pregnancy. In the early studies, where most patients received no pharmacological therapy for their IBD during pregnancy, it was hypothesized that this was due to elevated levels of plasma- I 7-hydroxycorticosteroids as pregnancy progressed (23). Since drug therapy is almost universal in more recent studies, there is insufficient data to separate the hormonal from pharmacological influences on disease activity. The role of psychological stress was also implied in older studies; “the mental attitude of a woman towards her pregnancy may be important in determining its effect on ulcerative colitis”(25).

As with all meta-analyses, caution needs to be used when making conclusions based on pooled studies with heterogeneous patient populations. In this context we have included studies that used possibly different methods for measuring “disease activity”, and included patients treated in very different decades. Correction for therapies used was not possible due to lack of sub-group data reported, although sensitivity analysis did not reveal a major difference between pre- or post- 1980 cohorts. To incorporate for statistical heterogeneity in the meta-analysis we used a random effects model to analyze all outcomes. We also performed sub-group and sensitivity analyses to incorporate for clinical heterogeneity, and examine differences in the overall effect estimate. It should also be noted that 13 of 14 included studies were judged to be at “high-risk” of bias in at least one quality domain (Suppl. Table 2).

In conclusion, this meta-analysis demonstrates that women who become pregnant when their IBD is active are more likely to experience disease activity during the pregnancy than those who enter pregnancy in clinical remission. This information should reinforce the importance of continuing maintenance medications in the pre-conception phase, and throughout pregnancy. Further data on relapse rates during pregnancy from modern, well-designed population-based cohorts would provide more meaningful data for clinical practice.

Supplementary Material

Supp Table S1-S2

Acknowledgments