Annals of Medicine

ISSN: (Print) (Online) Journal homepage: www.tandfonline.com/journals/iann20

Clinical characteristics and pregnancy outcomes in

pregnant women with TB: a retrospective cohort
study

Jiayu Wen & Jian-Qing He

To cite this article: Jiayu Wen & Jian-Qing He (2024) Clinical characteristics and pregnancy
outcomes in pregnant women with TB: a retrospective cohort study, Annals of Medicine, 56:1,
2401108, DOI: 10.1080/07853890.2024.2401108

To link to this article: https://doi.org/10.1080/07853890.2024.2401108

© 2024 The Author(s). Published by Informa

UK Limited, trading as Taylor & Francis
Group

View supplementary material

Published online: 13 Sep 2024.

Submit your article to this journal

Article views: 285

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=iann20
Annals of Medicine
2024, VOL. 56, NO. 1, 2401108
https://doi.org/10.1080/07853890.2024.2401108

Research Article

Clinical characteristics and pregnancy outcomes in pregnant women with

TB: a retrospective cohort study
Jiayu Wena,b and Jian-Qing Hea,b
a
Department of Respiratory and Critical Care Medicine, West China Hospital, Sichuan University, Chengdu, China; bState Key Laboratory
of Respiratory Health and Multimorbidity, West China Hospital, Sichuan University, Chengdu, China

ABSTRACT ARTICLE HISTORY

Purpose: The influence of pregnancy on tuberculosis (TB) has not been well studied. This study Received 27 September
aimed to investigate the demographics, clinical characteristics and outcomes of pregnant-related 2023
TB compared with the general population with TB. Revised 11 March 2024
Methods: We retrospectively analysed medical records of women during pregnancy or within six Accepted 19 April 2024
months postpartum with active TB who were admitted to the West China Hospital between 2011 KEYWORDS
and 2022. According to age, gender and admission time, the general population with active TB Pregnancy; miliary
was matched at a ratio of 1:2, and the demographics, clinical characteristics and outcomes were tuberculosis; tuberculosis;
compared. Mycobacterium
Results: All the participants in both the pregnant and non-pregnant groups were females, tuberculosis
averaging 26 years old, with a majority of Han nationality (72.4% vs. 69.5%, respectively). The two
groups were comparable (p < .05). Pregnant TB cases showed higher rates of fever (61% vs. 35%),
dyspnoea (39.9% vs. 18.7%), neurological symptoms (34.4% vs. 11.0%) and miliary TB (24.5% vs.
10.9%) compared to non-pregnant cases (p < .05). Additionally, the pregnant group exhibited
lower red blood cell counts (3.62 × 109/L vs. 4.37 × 109/L), lower albumin levels (31.20 g/L vs.
40.40 g/L) and elevated inflammatory markers (p < .05). Pregnant women with TB had severe
outcomes, with 16.3% requiring intensive care unit (ICU) care and a 3.3% TB-related mortality rate
– higher than local averages. In contrast, the non-pregnant group had lower rates (0.8% for ICU
admission, and no TB-related deaths). Moreover, active TB during pregnancies led to a high rate
of spontaneous abortion (34.1%), with military pulmonary TB identified as the sole risk factor for
severe TB in pregnancies (OR: 3.6; 95% CI: 1.15, 11.34).
Conclusions: Manifestations of TB in pregnant women differ from those in the general population
with TB. Pregnancy complicated with active TB greatly harms the mother and foetus and requires
special attention in the future.

Introduction pregnancies with active TB compared to pregnancies

without active TB [2]. Furthermore, infants born to
The true extent of tuberculosis (TB) during pregnancy
mothers with active TB exhibited a fourfold increased
remains uncertain; however, it is estimated that there
risk of perinatal mortality, highlighting the severity of
were approximately 216,500 cases of active TB world-
wide in 2011 [1]. The African region and the Southeast the consequences associated with TB during preg-
Asian region were identified as the regions with the nancy [2].
highest burdens, while China and Indonesia shared the Recent studies have demonstrated an elevated risk
fifth position in terms of TB prevalence [1]. of TB in pregnant women, including the increased risks
Active TB during pregnancy poses significant risks of maternal hospitalization and death and adverse
to both the mother and the infant [2–6]. A study pregnancy outcomes, primarily attributed to intricate
revealed that the odds of antenatal admission and alterations in the maternal immune system [2, 7–11].
miscarriage were found to be nine times greater in The presentation of TB during pregnancy is believed

CONTACT Jian-Qing He jianqing_he@scu.edu.cn Department of Respiratory and Critical Care Medicine, West China Hospital, Sichuan University,
Chengdu, Sichuan, China; State Key Laboratory of Respiratory Health and Multimorbidity, West China Hospital, Sichuan University, Chengdu, Sichuan, China
Supplemental data for this article can be accessed online at https://doi.org/10.1080/07853890.2024.2401108.
© 2024 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial License (http://creativecommons.org/licenses/by-nc/4.0/), which
permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited. The terms on which this article has been
published allow the posting of the Accepted Manuscript in a repository by the author(s) or with their consent.
2 J. WEN AND J.-Q. HE

to differ from the general population, leading to worse via telephone in January 2023 and obtained informed
outcomes. However, these differences have not been consent.
comprehensively described [12].
The objective of this study was to delineate the
demographics, clinical characteristics and short-term Definitions
outcomes of TB during pregnancy, in comparison to Factors predisposing to TB include the following:
the general population with TB. household contacts with exposure, human immuno-
deficiency virus (HIV) infection, diabetes mellitus,
chronic renal failure/haemodialysis, silicosis, gas-
Methods
trectomy [15–17], jejunoileal bypass [18, 19], hae-
Study population and data collection matological malignancies, head or neck carcinoma,
alcohol consumption and therapy with corticoste-
This retrospective cohort study enrolled patients
roids (prednisone or equivalent administered at a
diagnosed with active TB during pregnancy or within
dose of >15 mg/day for ≥1 month) [20].
six months (180 days) postpartum, who were admit-
Pregnancy is categorized into three stages: early
ted to West China Hospital between 2011 and 2022.
pregnancy (weeks 1–13), middle pregnancy (weeks
Using the Tenth Revision of the International
14–27) and late pregnancy (28th week until delivery).
Classification of Diseases (ICD-10) criteria set by the
Typical miliary TB occurs when Mycobacterium tuber-
World Health Organization (WHO), we retrieved data
culosis (MTB) spreads acutely through the lymphatic
on TB patients from electronic medical records (EMR).
and blood systems, resulting in the formation of small
Active TB cases included both bacteriologically con-
tubercles that resemble millet seeds in size and
firmed and clinically diagnosed cases [7]. Clinically
appearance (multiple 1–3-mm well-defined nodules
diagnosed cases were determined by TB specialists or
throughout all lung fields) [21,22].
respiratory physicians based on a combination of
The definitions of pulmonary tuberculosis (PTB), and
medical history, clinical signs and symptoms and
extrapulmonary tuberculosis (EPTB) were derived from
radiological changes suggestive of TB disease. The
WHO guidelines [23].
exclusion criteria were as follows: (1) a diagnosis of
Severe TB was defined as causing maternal death or
active TB before pregnancy, (2) pregnancy achieved
admission to the intensive care unit (ICU).
through in vitro fertilization–embryo transfer and (3)
incomplete clinical records. Pregnancy resulting from
in vitro fertilization–embryo transfer was not consid- Clinical outcome
ered due to evidence indicating that undergoing in
vitro fertilization–embryo transfer poses a possible The primary outcome of our study focused on mater-
risk for serious maternal problems and unfavourable nal mortality. Secondary outcomes included severe TB
perinatal outcomes in patients with TB during preg- and perinatal outcomes. For patients whose perinatal
nancy [13, 14]. In cases where a patient had multiple status remained unknown at discharge, a follow-up
admissions, only data from the first admission were was conducted via telephone in January 2023.
included. After manual screening, the remaining
patients were included in the study group (pregnant
Statistical analyses
group). Each case was matched with two
non-pregnant/non-postpartum patients with TB Categorical variables were presented as frequency
(non-pregnant group) based on age, gender and year with percentage and compared using the Chi-square
of admission time. or Fisher’s exact test. Continuous variables were
We extracted data for this study from medical reported as means with standard deviation (SD) or
records, which included demographics (such as age, median with interquartile range (IQR) and analysed
sex and ethnicity), symptoms, comorbidities, labora- using Student’s t-test or the Mann–Whitney U-test
tory findings, radiological presentations, diagnosis and (based on data distribution). Univariate and multi-
outcome. variate logistic regression were used to identify the
The study received ethical approval from the Ethics risk factors for severe disease. The significance level
Committee of West China Hospital, Sichuan University for the two-tailed p value was set at ≤.05. All statis-
(no. 761 (2019)) (including a waiver of informed con- tical analysis was performed using the SPSS Version
sent). For patients whose perinatal status remained 28.0 software (IBM, Chicago, IL), which was
unknown at discharge, a follow-up was conducted copyrighted.
 Annals of Medicine 3

Results Table 1. Baseline characteristics and symptomatology of preg-

nant group and non-pregnant group matched by age, gender
Study population and year of admission time.
Pregnant
From 2011 to 2022, a screening process based on the group Non-pregnant
ICD-10 codes from EMR identified a total of 51,669 Variables (n = 123) group (n = 246) p Value
admitted TB patients. Among them, 123 patients with Age (mean (SD)), years 26.63 (4.98) 26.69 (5.02) .916
Nationality .572
active TB were selected as the study group (pregnant Han nationality (n/%) 89 (72.4) 171(69.5)
group), while 246 patients were matched as the con-  Non-Han nationality 34 (27.6) 75(30.5)
trol group (non-pregnant group). The detailed flow- (n/%)
Household contact expose 5 (4.1) 7 (2.8) .756
chart is presented in Supplementary Figure 1. (n/%)
Complications
HIV infection (n/%) 0 (0) 1 (0.4) 1.000
 Chronic hepatopathy 9 (7.3) 9 (3.7) .124
Baseline characteristics and symptomatology (n/%)
 Diabetes mellitus (n/%) 1 (0.8) 1 (0.4) 1.000
For baseline characteristics, all participants in the study  Chronic renal failure 4 (3.3) 6 (2.4) .910
(n/%)
were female and the mean age in the pregnant group  Factors predisposing to 12 (9.8) 28 (11.4) .636
was 26.63 years, while in the non-pregnant group, it tuberculosisa (n/%)
Symptoms
was 26.69 years, with no statistically significant differ- Highest temperature <.001
ence observed (p = .916). Both groups were predomi- T > 38.5 °C (n/%) 67 (54.5) 66 (26.8)
nantly represented by individuals of Han nationality, 37.3 °C ≤ T ≤ 38.5 °C 8 (6.5) 20 (8.1)
(n/%)
and there were no significant disparities between T < 37.3 °C (n/%) 48 (39.0) 160 (65.0)
them (p = .572). Notably, household contact exposure  Constitutional 26 (21.1) 101 (41.1) <.001
symptomsb (n/%)
and complications, specifically factors predisposing to  Cough (n/%) 61 (49.6) 112 (45.5) .461
TB, exhibited no statistically significant variances  Expectoration (n/%) 40 (32.5) 74 (30.1) .633
Haemoptysis (n/%) 7 (5.7) 12 (4.9) .739
between the two groups, demonstrating comparability  Dyspnoea (n/%) 49 (39.9) 46 (18.7) <.001
(Table 1).  Chest pain (n/%) 22 (17.9) 64 (26.0) .082
 Neurological 30 (24.4) 27 (11.0) <.001
Regarding symptomatology, several noteworthy dif- symptomsc (n/%)
ferences emerged between the pregnant group and Gastrointestinal 16 (13.0) 38 (15.4) .532
the non-pregnant group. Among these, the natural symptomd (n/%)
TB: tuberculosis; SD: standard deviation; HIV: human immunodeficiency
pregnancy group exhibited higher frequencies of fever, virus; T: temperature. Bold values indicate p < .05 and are statistically
dyspnoea and neurological symptoms, while constitu- significant.
tional symptoms were more prevalent in the
a
Well-known factor predisposing to tuberculosis included: household con-
tacts exposed, HIV infection, diabetes mellitus, chronic renal failure/hae-
non-pregnant group. Fever emerged as the predomi- modialysis, silicosis, gastrectomy, jejunoileal bypass, haematological
nant clinical manifestation in the pregnant group, malignancies, head or neck carcinoma, alcohol consumption and therapy
with corticosteroids (prednisone or equivalent administered at >15 mg/day
accounting for 71% of total cases. The further stratified for ≥1 month).
analysis revealed that cases with high fever (highest b
Constitutional symptoms include weight loss, fatigue and/or night sweats.
c
Neurological symptoms include headache, disturbance of consciousness,
body temperature >38.5 °C) constituted 54.5% of the impaired physical activity and speech disorder.
total cases. On the other hand, constitutional symp- d
Gastrointestinal symptom include abdominal pain, diarrhoea, nausea and
vomiting.
toms were most frequently observed in the
non-pregnant group (Table 1).
non-pregnant group in terms of lung involvement. The
pregnant group displayed a larger extent of lung
involvement, with a higher prevalence of typical mili-
Radiological and laboratory findings
ary TB and pleural effusion compared to the
Due to concerns regarding radiation exposure during non-pregnant group. However, cavities were less fre-
pregnancy, a proportion of patients in the pregnant quently detected in the group of pregnant TB individ-
group declined to undergo relevant examinations. uals (Table 2).
Only 86.2% of patients in this group received chest In terms of routine laboratory investigations, the
imaging, while the non-pregnant group had a higher pregnant group exhibited lower levels of red blood
completion rate of 96.7% for chest imaging, leading to cell (RBC) count and albumin compared to the
a significant difference between the two groups non-pregnant group. Conversely, the white blood cell
(p < .001). Among the subset of individuals who (WBC) count, neutrophil-to-lymphocyte ratio (NLR) and
underwent complete chest imaging, notable disparities C-reactive protein (CRP) levels were higher in the preg-
were observed between the pregnant group and the nant group as compared to the non-pregnant group.
4 J. WEN AND J.-Q. HE

Table 2. Radiological and laboratory findings of pregnant

group and non-pregnant group matched by age, gender and
year of admission time.
Pregnant group Non-pregnant
Variables (n = 123) group (n = 246) p Value
Chest imagings 106 (86.2) 238 (96.7) <.001
performeda (n/%)
Radiological findings .036
Bilateral pulmonary 69 (65.1) 116 (48.7)
infiltratesb (n/%)
Unilateral 16 (15.1) 59 (24.8)
pulmonary
infiltratesb (n/%)
 Noneb (n/%) 21 (19.8) 63 (26.5)
Typical miliary TBb 26 (24.5) 26 (10.9) .033
(n/%)
 Cavitary infiltratesb 4 (3.7) 42 (17.6) <.001
(n/%)
Pleural effusionb 52 (49.1) 71 (29.0) <.001
(n/%)
Regular laboratory
findings
RBC (median (IQR)), 3.62 (3.21, 4.37 (3.99, 4.71) <.001
×109/L 4.06)
PLT (median (IQR)), 264.00 (193.00, 249.00 (185.75, .075
×109/L 378.00) 320.00)
WBC (median 7.37 (5.35, 6.14 (5.35, 10.01) <.001
(IQR)), ×109/L 10.01)
 NLR (median (IQR)) 7.71 (4.85, 3.29 (2.19, 5.41) <.001
12.45)
 CRP (median (IQR)), 61.40 (20.92, 9.31 (3.95, 33.90) <.001 Figure 1. Information related to TB diagnosis. (A) The specific
mg/L 92.63) distribution of TB between the pregnant group and the
 Serum albumin 31.20 (27.30, 40.40 (35.80, <.001
(median (IQR)), 35.80) 43.40)
non-pregnant group (*p < .05, **p < .01); (B) Onset time dis-
g/L tribution of the pregnant group.
Immunological tests
for tuberculosis
 IGRA positivec (n/%) 48/68 (70.6) 89/111 (80.2) .142
differences emerged. Lymphatic TB was the most com-
TST positived (n/%) 51/86 (59.3) 122/158 (77.2) .003 mon form in the non-pregnant group, while tubercu-
TB antibody (n/%)
e
19/65 (29.2) 51/115 (44.3) .046 lous pleuritis and tuberculous meningitis predominated
TB: tuberculosis; IQR: interquartile range; RBC: red blood count; WBC:
white blood count; PLT: platelets count; NLR: neutrophil-to-lymphocyte
in the pregnant group (Figure 1(A)). In our study, the
ratio; CRP: C-reactive protein; IGRA: interferon gamma release assay; TST: pregnant group had a shorter median time from
tuberculin skin test. Bold values indicate p < .05 and are statistically symptom onset to diagnosis compared to the
significant.
a
Chest imagings including X-ray, computed tomography and magnetic non-pregnant group. Although there was no statisti-
resonance imaging. cally significant difference in the utilization of invasive
b
Among those with chest imagings performed.
c
Among those with IGRA performed. diagnostic procedures between the two groups, it is
d
Among those with TST performed. important to note that the pregnant group still had a
e
Among those with TB antibody performed.
lower rate of bacteriologic confirmation compared to
the non-pregnant group (Supplementary Table S1).
Regarding auxiliary diagnostic immunological tests for In the pregnant group, 48.8% of patients experi-
TB, including the tuberculosis interferon-gamma enced TB-related clinical symptoms during the middle
release assay (TB-IGRA), tuberculin skin test (TST) and stage of pregnancy, as indicated by the detailed distri-
TB antibodies, both the pregnant group and bution provided in Figure 1(B).
non-pregnant group demonstrated the highest sensi-
tivity with TB-IGRA, followed by the TST, while TB anti-
Outcome
bodies displayed the lowest sensitivity (Table 2).
Within the pregnant group, the rate of ICU admission for
pregnant women was 16.3%, a substantial increase com-
Diagnosis
pared to the rate of 0.8% observed in the non-pregnant
The distribution of TB between the pregnant group group. Moreover, the TB-related mortality rate stood at
and the non-pregnant group showed similar propor- 3.3% in the pregnant group, while no deaths were
tions of PTB, with 82.9% and 77.6%, respectively, and reported in the non-pregnant group. These findings high-
no statistically significant difference was observed light a heightened susceptibility of pregnant women with
(p = .237). However, when considering EPTB, notable TB to ICU admission and TB-related mortality, with a
 Annals of Medicine 5

Table 3. Outcome of pregnant group and non-pregnant group Table 4. Multiple logistic regression analysis of risk factors for
matched by age, gender and year of admission time. severe pregnant group.
Pregnant group Non-pregnant Risk factors OR (95% CI) p Value
Variables (n = 123) group (n = 246) p Value Haemoglobin 0.99 (0.95, 1.02) .396
Maternal outcome WBC 1.15 (0.98, 1.34) 1.114
TB-related mortality 4 (3.3) 0 (0) .012 Typical miliary TB 3.60 (1.15, 11.34) .028
(n/%) TB: tuberculosis; OR: odds ratio; CI: confidence interval; WBC: white blood
Admission to the ICU 20 (16.3) 2 (0.8) <.001 count. Bold values indicate p < .05 and are statistically significant.
(n/%)
Perinatal outcome
 Spontaneous 42a (34.1)
miscarriage (n/%) assessment, no statistically significant differences
 Induced abortion 29b (23.6)
(n/%) were observed at baseline (age, nationality and
Preterm birth (n/%) 5 (4.1) Charlson score) between the bacteriologically con-
 Normal delivery 47 (38.2)
(n/%)
firmed TB group and the clinically diagnosed TB
TB: tuberculosis. Bold values indicate p < .05 and are statistically group, indicating comparability. No statistically signif-
significant. icant distinctions emerged when evaluating TB-related
a
Among them, 18 experienced a spontaneous miscarriage in early preg-
nancy, 23 in mid-pregnancy and one in late pregnancy.
maternal mortality and perinatal outcomes between
b
Among them, 24 had an abortion because of concerns about tuberculosis the two groups (Supplementary Table S2).
or drug side effects on the foetus, and only five had an abortion because
of inevitable abortion.

Risk factors for severe pregnant group

statistically significant disparity between the two groups The severe pregnant group was defined pregnant
(p < .05) (Table 3). According to epidemiological investi- women who required ICU admission for supportive
gations, Sichuan Province has experienced a significant treatment or died. Among the study participants, 20
decrease in maternal mortality, declining from 37.16 per individuals were classified as severe TB, while 103 were
100,000 in 2011 to 13.09 per 100,000 in 2022 [24,25]. categorized as non-severe TB. Utilizing univariate logis-
However, in our study, when compared to the lowest tic regression analysis, we examined potential risk fac-
maternal mortality rates recorded in 2022, our rates were tors and identified statistically significant associations
considerably higher than the average. (p < .05) with haemoglobin levels, WBC counts and
In the pregnant group, perinatal outcomes raised typical miliary TB on imaging (Supplementary Table
significant concerns, particularly due to a markedly S3). The three variables were subsequently included in
elevated spontaneous miscarriage rate of 34.1%. a multivariate logistic regression analysis. In this analy-
Examination of gestational age distribution revealed sis, only the presence of typical miliary TB on imaging
that 18 miscarriages occurred during early pregnancy, remained significantly associated as an independent
constituting 64.3% of the total in this stage. For risk factor for the severe pregnant group (OR: 3.6; 95%
mid-pregnancy, 23 miscarriages were observed, CI: 1.15, 11.34) (Table 4).
accounting for 38.3% of the total in this phase. Late
pregnancy witnessed one miscarriage, representing
Discussion
5.3% of the total occurrences in this trimester. TB in
early pregnancy has a great adverse impact on the In our study, we sought to compare patients hospital-
foetus. Induced abortions constituted 23.6% of the ized with pregnancy-related active TB to the hospital-
overall cases. Moreover, the preterm birth rate was ized general population with active TB spanning the
4.1%, while normal deliveries occurred in 38.2% of years 2011–2022. Our analysis revealed significant dif-
cases. Among the 29 cases where induced abortion ferences across various parameters between the two
was chosen, 24 individuals (19.5% of the total) made groups, including clinical manifestations, laboratory
this decision due to concerns regarding the potential findings, imaging characteristics and outcomes.
impact of TB or medication side effects on the foetus. Pregnancy-related TB displayed more typical clinical
Only five individuals opted for induced abortion due presentations, such as fever, dyspnoea and neurologi-
to unavoidable miscarriage. These findings underscore cal symptoms, with a higher prevalence of miliary TB.
the complex and challenging perinatal outcomes in Pregnant patients displayed significantly lower RBC
the context of TB during pregnancies (Table 3). and albumin counts, coupled with elevated levels of
Then, we conducted an exploratory comparison of inflammatory markers. This suggested a more pro-
outcomes among pregnant women with bacteriolog- nounced inflammatory response and poorer nutritional
ically and clinically diagnosed TB. In the initial status in the pregnancy group. Throughout the
6 J. WEN AND J.-Q. HE

hospitalization period, pregnant patients with TB expe- decrease in IFN-γ secretion is observed due to the
rienced notably higher rates of ICU admission and impact on Th1, which contributes to an increased sus-
TB-related mortality compared to their non-pregnant ceptibility to MTB infection and dissemination
counterparts. In conclusion, considering all these [26–29,34,35].
observed differences, our study indicated that preg- The possibility of TB should be considered when
nancy might contribute to the dissemination and exac- clinical symptoms such as fever occur during preg-
erbation of TB. Besides, it is noteworthy that the nancy in areas where TB is endemic. The imaging and
spontaneous abortion rate in pregnancies complicated immunological and etiological examinations of TB
by active TB is as high as 34.1%, particularly in early should be completed. According to the current guide-
pregnancy when the rate is the highest. Additionally, lines, the radiation exposure dose of X-ray and CT
apart from the high spontaneous abortion rate, our examination is much lower than the exposure dose
study revealed that 19.5% of pregnant women opted associated with foetal injury, and pregnant women
for induced abortion due to concerns about TB and should be encouraged to receive necessary imaging
the potential adverse effects of anti-TB drugs. Finally, examinations, because the risk of misdiagnosis and
we found the sole independent predictor of severe TB delayed diagnosis is much higher than the risk of radi-
cases in pregnancies was the presence of typical mili- ation [36]. Compared with TST and TB antibodies,
ary TB findings on imaging (OR: 3.6; 95% CI: 1.15, 11.34). TB-IGRA demonstrated higher sensitivity during preg-
Our study indicated that pregnancy may contribute nancy in TB-endemic countries, consistent with previ-
to the dissemination and exacerbation of TB, which ous studies [37–39].
contradicts the findings of a recent large-scale In our study, out of the 123 individuals with
case-control study [9]. The discrepancy in results may pregnant-related TB, 24 individuals (19.5%) opted
be attributed to variations in the definitions of TB for induced abortion due to concerns about the
exacerbations used in our respective studies. The potential impact of TB or drug side effects on the
case-control study defined severe TB solely based on foetus. Current research indicates that untreated TB
the presence of at least two criteria, such as multi-organ can have significant effects on pregnant women and
involvement, positive culture results or hospitalization their unborn babies. If the patient was contracting
exceeding 10 days. According to their findings, preg- TB, anti-TB treatment should be initiated [40–42].
nancy was not identified as a contributing factor to With proper treatment, pregnant women with TB
the exacerbation of TB. However, it is important to disease can have successful outcomes [43]. For
note that their definition of severe TB was overly drug-sensitive TB, a standard four-drug regimen is
broad and incomplete. In contrast, our study utilized recommended, including isoniazid (INH), rifampicin
more specific and comprehensive clinical data obtained (RIF), pyrazinamide (PZA) and ethambutol (EMB)
from healthcare institutions. This allowed us to thor- [44,45]. Although there is currently a lack of
oughly examine the impact of pregnancy on TB from high-quality safety studies on PZA in pregnant
multiple perspectives, thereby enhancing the credibil- women, potential risks cannot be completely ruled
ity of our findings. out [46]. Therefore, the use of PZA should be
Pregnancy can contribute to the worsening and decided on a case-by-case basis, taking into consid-
spread of TB, which may be mediated by changes in eration the patient’s preferences and the severity of
immune function caused by hormone fluctuations the disease. However, INH, RIF and EMB are gener-
during pregnancy [26–31]. Compared to the ally considered to be relatively safe [47]. For diag-
non-pregnant group, the pregnant group exhibited a nosed and treated TB cases with relatively safe
significant decrease in positive rates of TST and TB medications, overly aggressive measures such as
antibodies (p < .05), which also reflected the effect of induced abortion may not be necessary. TB special-
pregnancy on immune function [21]. During preg- ists or infectious disease physicians should
nancy, levels of oestrogen and progesterone signifi- strengthen patient education and actively monitor
cantly increase. Those sex hormones can lead to a for drug-related adverse effects.
decrease in type 1 helper T cell (Th1) mediated
immune responses and CD4+ T cell proliferation [32,33].
Limitations and recommendations
Interferon-gamma (IFN-γ), an important cytokine
involved in controlling the infection and spread of Our study is subject to several limitations that should
MTB, plays a role in granuloma formation and the gen- be acknowledged. First, it is a study conducted at a
eration of reactive oxygen species. Its secretion single centre, which may restrict the ­generalizability of
depends on Th1 cells. Therefore, during pregnancy, a our findings to other populations or settings. As a
 Annals of Medicine 7

retrospective study, patient characteristics were telephone in January 2023 and obtained informed consent.
retrieved from the hospital’s database, resulting in The procedures used in this study adhere to the tenets of
the Declaration of Helsinki.
potential bias due to loss of data variable capture and
shifting of recall. Secondly, there is a potential for
selection bias as our analysis focused solely on Disclosure statement
hospitalized patients, which may not represent the
­
No potential conflict of interest was reported by the
entire spectrum of pregnancy-related TB cases. It may
author(s).
also overestimate foetal and maternal mortality. Thirdly,
the study duration spans a 10-year period, during
which there have been substantial advancements in Funding
disease management practices, data collection
No funding was received to assist with the preparation of
­methodologies and socioeconomic factors. For instance, this manuscript.
the ­implementation of TB-IGRA testing in our hospital
was initiated only in 2012, leading to some missing
data. Nevertheless, we made efforts to minimize this Data availability statement
bias through meticulous matching of patients based The datasets used and/or analysed during the current study
on their admission time. Lastly, our study lacks data on are available from the corresponding author upon reason-
the incidence of pregnancy-related TB, the prevalence able request.
of latent TB infection before pregnancy and foetal
mortality with normal pregnancy cases in our hospital
or region. It is challenging to distinguish between References
reactivation of latent TB or new infections as the cause [1] Sugarman J, Colvin C, Moran AC, et al. Tuberculosis in
of newly diagnosed active TB cases. Furthermore, the pregnancy: an estimate of the global burden of dis-
impact of active TB during pregnancy on foetal out- ease. Lancet Glob Health. 2014;2(12):e710–e716. doi:
10.1016/S2214-109X(14)70330-4.
comes could not be comprehensively evaluated. To
[2] Sobhy S, Babiker Z, Zamora J, et al. Maternal and peri-
address these limitations, future research endeavours natal mortality and morbidity associated with tubercu-
should aim for multicentre prospective study designs losis during pregnancy and the postpartum period: a
with standardized data collection protocols. Place the systematic review and meta-analysis. BJOG. 2017;124(5):
screening and comprehensive management of TB in 727–733. doi: 10.1111/1471-0528.14408.
[3] Nguyen Y, McNabb KC, Farley JE, et al. Examining fam-
women of childbearing age on the agenda.
ily planning and adverse pregnancy outcomes for
women with active tuberculosis disease: a systematic
review. BMJ Open. 2022;12(3):e054833. doi: 10.1136/
Conclusions bmjopen-2021-054833.
In summary, patients with TB related to pregnancy [4] Dennis EM, Hao Y, Tamambang M, et al. Tuberculosis
during pregnancy in the United States: racial/ethnic
exhibit more severe manifestations and worse progno-
disparities in pregnancy complications and in-hospital
sis than the general population. Pregnancy compli- death. PLOS One. 2018;13(3):e0194836. doi: 10.1371/
cated with active TB greatly is harmful to the mother journal.pone.0194836.
and foetus, it is imperative to prioritize and enhance [5] Sade S, Wainstock T, Sheiner E, et al. Perinatal outcome
screening and prevention efforts in this unique popu- and long-term infectious morbidity of offspring born to
women with known tuberculosis. J Clin Med. 2020;9(9):
lation to minimize maternal and child health impact.
2768. doi: 10.3390/jcm9092768.
[6] Menéndez C, Romagosa C, Ismail MR, et al. An autopsy
study of maternal mortality in Mozambique: the contri-
Author contributions bution of infectious diseases. PLoS Med. 2008;5(2):e44.
Jiayu Wen: article design, data collection, writing. Jian-Qing doi: 10.1371/journal.pmed.0050044.
He: article design, project administration, writing – review [7] Zenner D, Kruijshaar ME, Andrews N, et al. Risk of tu-
and editing. berculosis in pregnancy: a national, primary care-based
cohort and self-controlled case series study. Am J
Respir Crit Care Med. 2012;185(7):779–784. doi: 10.1164/
Ethical approval rccm.201106-1083OC.
[8] Jonsson J, Kühlmann-Berenzon S, Berggren I, et al.
The study received ethical approval from the Ethics Increased risk of active tuberculosis during pregnancy
Committee of West China Hospital, Sichuan University [No. and postpartum: a register-based cohort study in
761 (2019)]. For patients whose perinatal status remained Sweden. Eur Respir J. 2020;55(3):1901886. doi: 10.1183/
unknown at discharge, a follow-up was conducted via 13993003.01886-2019.
8 J. WEN AND J.-Q. HE

[9] Nordholm AC, Suppli CH, Norman A, et al. Pregnancy [24] Statistical Communiqué of Sichuan National Economic
and post-partum tuberculosis; a nationwide register-based and Social Development in 2011. Sichuan Provincial
case-control study, Denmark, 1990 to 2018. Euro Surveill. Bureau of Statistics; 2012 [cited 2023 Dec 30]. Available
2022;27(12):2100949. doi: 10.2807/1560-7917.ES.2022.27. from: https://tjj.sc.gov.cn/scstjj/c105897/2012/4/6/76688
12.2100949. a256e064ce28aaf8ccded42de69.shtml
[10] Salazar-Austin N, Hoffmann J, Cohn S, et al. Poor ob- [25] Report on the health status and key diseases of the popu-
stetric and infant outcomes in human immunodeficien- lation in Sichuan Province in 2022. Sichuan Provincial Health
cy virus-infected pregnant women with tuberculosis in Commission; 2023 [cited 2023 Dec 30]. Available from:
South Africa: the Tshepiso study. Clin Infect Dis. https://wsjkw.sc.gov.cn/scwsjkw/qtwj/gfxwj_new2.shtml
2018;66(6):921–929. doi: 10.1093/cid/cix851. [26] Krishnan L, Guilbert LJ, Russell AS, et al. Pregnancy im-
[11] Jones AJ, Mathad JS, Dooley KE, et al. Evidence for im- pairs resistance of C57BL/6 mice to Leishmania major
plementation: management of TB in HIV and pregnan- infection and causes decreased antigen-specific
cy. Curr HIV/AIDS Rep. 2022;19(6):455–470. doi: 10.1007/ IFN-gamma response and increased production of T
s11904-022-00641-x. helper 2 cytokines. J Immunol. 1996;156(2):644–652. doi:
[12] Stewart DA. Pregnancy and tuberculosis: part I—the ef- 10.4049/jimmunol.156.2.644.
fects of pregnancy on tuberculosis. Can Med Assoc J. [27] Flynn JL, Chan J, Triebold KJ, et al. An essential role for
1922;12(1):1–3. interferon gamma in resistance to Mycobacterium tuber-
[13] Xia L, Mijiti P, Liu XH, et al. Association of in vitro fertil- culosis infection. J Exp Med. 1993;178(6):2249–2254.
ization with maternal and perinatal outcomes among doi: 10.1084/jem.178.6.2249.
pregnant women with active tuberculosis: a retrospec- [28] Matta SK, Kumar D. Hypoxia and classical activation lim-
tive hospital-based cohort study. Front Public Health. its Mycobacterium tuberculosis survival by Akt-dependent
2022;10:1021998. doi: 10.3389/fpubh.2022.1021998. glycolytic shift in macrophages. Cell Death Discov.
[14] Dong S, Zhou R, Peng E, et al. Analysis of clinical fea- 2016;2(1):16022. doi: 10.1038/cddiscovery.2016.22.
tures and risk factors in pregnant women with miliary [29] Sasindran SJ, Torrelles JB. Mycobacterium tuberculosis in-
pulmonary tuberculosis after in vitro fertilization em- fection and inflammation: what is beneficial for the
bryo transfer. Front Cell Infect Microbiol. 2022;12:885865. host and for the bacterium? Front Microbiol. 2011;2:2.
doi: 10.3389/fcimb.2022.885865. doi: 10.3389/fmicb.2011.00002.
[15] Park HY, Choi SH, Kim D, et al. Incidence and risk factors [30] Kovats S, Carreras E, Agrawal H. Sex steroid receptors in
of tuberculosis in patients following gastrectomy or en- immune cells. In: Klein S & Roberts C, editors. Sex hor-
doscopic submucosal dissection: a cohort analysis of mones and immunity to infection; 2010. Berlin,
country-level data. Gastric Cancer. 2023;26(3):405–414. Heidelberg: Springer.
doi: 10.1007/s10120-023-01367-4. [31] Sharma S, Rodrigues PRS, Zaher S, et al. Immune-metabolic
[16] Jung WJ, Park YM, Song JH, et al. Risk factors for tuber- adaptations in pregnancy: a potential stepping-stone to
culosis after gastrectomy in gastric cancer. World J sepsis. EBioMedicine. 2022;86:104337. doi: 10.1016/j.ebi-
Gastroenterol. 2016;22(8):2585–2591. doi: 10.3748/wjg. om.2022.104337.
v22.i8.2585. [32] Piccinni MP, Raghupathy R, Saito S, et al. Cytokines,
[17] Cheng KC, Liao KF, Lin CL, et al. Gastrectomy correlates hormones and cellular regulatory mechanisms favoring
with increased risk of pulmonary tuberculosis: a successful reproduction. Front Immunol. 2021;12:717808.
population-based cohort study in Taiwan. Medicine. doi: 10.3389/fimmu.2021.717808.
2018;97(27):e11388. doi: 10.1097/MD.0000000000011388. [33] Lissauer D, Eldershaw SA, Inman CF, et al. Progesterone
[18] Snider DEJr. Jejunoileal bypass for obesity: a risk factor promotes maternal–fetal tolerance by reducing human
for tuberculosis. Chest. 1982;81(5):531–532. doi: 10.1378/ maternal T-cell polyfunctionality and inducing a specific
chest.81.5.531. cytokine profile. Eur J Immunol. 2015;45(10):2858–2872.
[19] Bruce RM, Wise L. Tuberculosis after jejunoileal bypass doi: 10.1002/eji.201445404.
for obesity. Ann Intern Med. 1977;87(5):574–576. doi: [34] Bhosale R, Alexander M, Deshpande P, et al. Stages of
10.7326/0003-4819-87-5-574. pregnancy and HIV affect diagnosis of tuberculosis in-
[20] Benito N, García-Vázquez E, Horcajada JP, et al. Clinical fection and Mycobacterium tuberculosis (MTB)-induced
features and outcomes of tuberculosis in transplant re- immune response: findings from PRACHITi, a cohort
cipients as compared with the general population: a study in Pune, India. Int J Infect Dis. 2021;112:205–211.
retrospective matched cohort study. Clin Microbiol doi: 10.1016/j.ijid.2021.09.010.
Infect. 2015;21(7):651–658. doi: 10.1016/j.cmi.2015.03.010. [35] Mathad JS, Bhosale R, Balasubramanian U, et al.
[21] Sharma SK, Mohan A. Miliary tuberculosis. Microbiol Quantitative IFN-γ and IL-2 response associated with la-
Spectr. 2017;5(2). doi: 10.1128/microbiolspec.TNMI7- tent tuberculosis test discordance in HIV-infected preg-
0013-2016. nant women. Am J Respir Crit Care Med. 2016;193(12):1421–
[22] Mert A, Bilir M, Tabak F, et al. Miliary tuberculosis: clini- 1428. doi: 10.1164/rccm.201508-1595OC.
cal manifestations, diagnosis and outcome in 38 adults. [36] Committee Opinion No. 723: guidelines for diagnostic
Respirology. 2001;6(3):217–224. doi: 10.1046/j.1440- imaging during pregnancy and lactation. Obstet
1843.2001.00328.x. Gynecol. 2017;130(4):e210–e216.
[23] World Health Organization. Definitions and reporting [37] Escudero JN, Mecha J, Richardson BA, et al. Impact of hu-
framework for tuberculosis – 2013 revision: updated man immunodeficiency virus and peripartum period on
December 2014 and January 2020. Geneva, Switzerland: Mycobacterium tuberculosis infection detection. J Infect
World Health Organization; 2013. Dis. 2023;228(12):1709–1719. doi: 10.1093/infdis/jiad416.
 Annals of Medicine 9

[38] Mathad JS, Bhosale R, Sangar V, et al. Pregnancy differen- nancy: a prospective cohort study. BMC Infect Dis.
tially impacts performance of latent tuberculosis diagnos- 2020;20(1):686. doi: 10.1186/s12879-020-05416-6.
tics in a high-burden setting. PLOS One. 2014;9(3):e92308. [44] Czeizel AE, Rockenbauer M, Olsen J, et al. A
doi: 10.1371/journal.pone.0092308. population-based case-control study of the safety of oral
[39] LaCourse SM, Cranmer LM, Matemo D, et al. Effect of anti-tuberculosis drug treatment during pregnancy. Int J
pregnancy on interferon gamma release assay and tuber- Tuberc Lung Dis. 2001;5(6):564–568.
culin skin test detection of latent TB infection among [45] World Health Organization. Tuberculosis: WHO global tu-
HIV-infected women in a high burden setting. J Acquir berculosis report 2014. Geneva: WHO; 2014.
Immune Defic Syndr. 2017;75(1):128–136. doi: 10.1097/ [46] Kaplan JE, Benson C, Holmes KK, et al. Guidelines for
QAI.0000000000001298.
prevention and treatment of opportunistic infections in
[40] Davidson PT. Managing tuberculosis during pregnancy.
HIV-infected adults and adolescents: recommendations
Lancet. 1995;346(8969):199–200. doi: 10.1016/s0140-6736
from CDC, the National Institutes of Health, and the HIV
(95)91263-0.
[41] Snider DEJr., Layde PM, Johnson MW, et al. Treatment Medicine Association of the Infectious Diseases Society
of tuberculosis during pregnancy. Am Rev Respir Dis. of America. MMWR Recomm Rep. 2009;58(Rr-4):1–207;
1980;122(1):65–79. doi: 10.1164/arrd.1980.122.1.65. quiz CE201-204.
[42] Jana N, Vasishta K, Jindal SK, et al. Perinatal outcome in [47] Nahid P, Dorman SE, Alipanah N, et al. Official American
pregnancies complicated by pulmonary tuberculosis. Int Thoracic Society/Centers for Disease Control and
J Gynaecol Obstet. 1994;44(2):119–124. doi: 10.1016/0020- Prevention/Infectious Diseases Society of America Clinical
7292(94)90064-7. Practice Guidelines: treatment of drug-susceptible tuber-
[43] van de Water BJ, Brooks MB, Huang CC, et al. Tuberculosis culosis. Clin Infect Dis. 2016;63(7):e147–e195. doi: 10.1093/
clinical presentation and treatment outcomes in preg- cid/ciw376.