Tuberculosis Care for Pregnant Women
Tuberculosis Care for Pregnant Women
Thirty five studies were selected for detailed review. Good inter-rater agreement on the quality assessment of the studies was found (K = 0.70). There were 6 studies with good quality (++), 21 with medium quality (+), and 8 with low quality (−). The reasons to include studies with low quality is the limited number of studies in selected topics and one of the purpose of this review is to have an overall evaluation on the situation of research on TB care for pregnant women.
Table 1 shows the characteristics of the reviewed studies and the quality evaluation scores. Concerning the study types, there were 14 cohort studies, 8 before and after studies, 6 cross sectional studies, 4 case control studies, 2 controlled trials, and 1 qualitative study.
Concerning the topics addressed, some studies addressed multiple topics: 8 studies covered 2 topics and 1 study covered 3. The topic of TB diagnosis/screening was presented in 21 studies, 14 studies addressed TB treatment (including 4 on MDR treatment), 2 studies were on TB prevention, specifically on LTBI prophylaxis, and 3 studies were on the follow-up of women's compliance to the TB prophylaxis therapy. All studies were published after 1975, with a majority (26/35) published after 1999. Data originated from 16 different countries. Fifteen studies were carried out in resource-rich countries: US (11 studies), UK (3), and the Netherlands (1), while 20 were carried out in resource-limited countries: South Africa (6 studies), India (3), and the remaining countries had 1 each. One group of researchers in South Africa conducted 3 studies, other studies were all conducted by different groups of researchers. In all, 81093 people were enrolled in these studies, 37404 of whom in the study group (pregnant women with TB) and 43689 in the controlled/comparison groups (pregnant women without TB or/and non-pregnant women with TB). The large total was mostly due to one population based study involving 61016 people.
The procedures for TB screening and diagnosis for pregnant women described in the reviewed studies include the tuberculin sensitivity test (TST/PPD) followed by the sputum test (Acid-fast bacillus - AFB) and the shielded chest X-ray. The AFB smear test appears to have low sensitivity in pregnant women, but is still used in low resource settings as part of the procedure for diagnosing active TB due to its low cost and simple technique. AFB culture was used as a confirmation of diagnosis, but is time consuming and not available in low resource settings. Another technique, fluorescence microscopy, was recommended as a substitute for AFB culture because it is cheaper. This procedure is for pulmonary TB cases and cannot identify extra pulmonary TB without additional tests and presence of physical TB symptoms. A clinical examination comprising a questionnaire tracking TB history and detecting TB clinical symptoms was also used and was proven to increase reliability of TB screening and diagnosis when combined with paraclinical tests. Some authors recommended not using chest X-ray for pregnant women if there were no clinical symptoms of TB.
TST is used widely as the first step in TB screening and diagnosis and to identify LTBI. Studies showed that pregnancy does not affect the sensitivity of this test, but its result can be affected by HIV infection or any situation that severely weakens the immune system (such as disseminated TB), as these could lead to false negative results. BCG vaccination can also lead to TST positive results in healthy women. In a high HIV prevalence setting, other tests and clinical symptoms should therefore be taken into account in diagnosing TB and the TST and anergy skin tests (the latter is used to evaluate whether the immune system is functioning properly or not and can indicate whether the results of the other skin test are reliable) are recommended as a TB screening method in the prenatal care procedures. In populations in which the majority of people are BCG vaccinated or their vaccination status is uncertain, TST is discouraged and IGRA is recommended for TB screening and diagnosis.
Concerning the IGRA test, one study in Kenya compared results of this test with the TST in screening for TB and showed the advantage of the IGRA test over TST in TB screening and diagnosis for HIV positive pregnant women, since its sensitivity is not affected by HIV infection. Two studies proved the value of IGRA in detecting LTBI in pregnant women, since the results of this test were not affected by BCG vaccination, thus avoiding the TST false positive result and the unnecessary, consequent INH prophylaxis.
TB prevention includes BCG vaccination in childhood and INH prophylaxis for LTBI positive people. There were 2 studies on TB prevention and both were on INH prophylaxis for LTBI pregnant women. Both were conducted in the US with pregnant women of foreign origin and LTBI was diagnosed by TST. One study showed a low completion rate of INH therapy ( 9.3%) and the other showed a high risk of INH toxic hepatitis, with pregnant women having a 2.5 fold greater risk of INH hepatitis than non-pregnant women (but this result was not statistically significant due to the small number of women). The 2 studies found that the main reason for this discouraging result was a lack of follow up and referral services for pregnant women undergoing INH prophylaxis.
There were 375 pregnant women with TB in the 14 studies on TB treatment. Treatment outcome was generally positive, with 332/375 women cured (confirmed by AFB culture conversion). In terms of mortality, 25 women died during treatment, 11 of whom died due to meningitis TB, 11 due to MDR-TB, 2 due to acute respiratory distress syndrome (ARDS), and 1 due to a non-TB related reason (massive pulmonary embolism). Other than mortality, the negative treatment outcomes included 4 treatment failures, 4 cases of residual functional deficit, 7 treatment terminations due to adverse drug effects, and 3 cases of treatment abandonment.
In terms of pregnancy outcome, only 11 women chose to terminate the pregnancy when they discovered their TB situation, while the others continued the pregnancy and underwent TB therapy. Among the pregnant women undergoing TB therapy, 332 women gave birth, 1 had a therapeutic abortion, 3 had miscarriages, 3 had stillborns, and 25 died. Concerning the 332 cases of mothers who gave birth, 4 infants died shortly after birth due to pneumonia and prematurity, 2 were HIV positive, 1 had active TB, 2 had LTBI, 50 were low birth weight, and 7 had growth restriction. Studies also showed that HIV infected women were more likely to choose pregnancy termination and had higher mortality and morbidity rates, even with intensive TB treatment combined with HIV treatment.
The first line drugs used for pregnant women in the studies included INH, ethambutol (ETB), rifampicin (RIF), and, in some cases of extra pulmonary TB, pyrazinamid (PZA). In the MDR cases, the second line drugs, including drugs of the amino glycosides group, fluoroquinolone, thioamides, cycloserines, and terizidone, were used in combination with effective first line drugs, and the treatment regime depended on the drug resistance situation of the individual cases.
Regarding effectiveness and safety of anti-TB drugs, results of the reviewed studies showed no significant association between child abnormality and mother's exposure to anti-TB drugs, both for 1 and 2 line anti-TB drugs during pregnancy. Other significant adverse effects were recorded in a very small number of pregnant women (2 cases of drug induced hepatitis, 2 of PZA allergy, 2 of sensorineural deafness, and 1 of severe nausea and jaundice) and led to termination of therapy without mortality. Streptomycin (SM) was not used in any studies because of its potential risk of deafness in babies. However, a retrospective study was conducted, checking the hearing capacity of 30 children whose mothers received SM injection during pregnancy, and found no significant effect, with only one case of deafness possibly linked to the mother's use of SM. Authors of this study recommend only using SM after the 2 trimester if really necessary.
There were 4 studies with 55 pregnant women on MDR treatment. Unlike other studies, in these TB was detected in all the women before pregnancy. More specifically, 48/55 of the women had been diagnosed with MDR TB and had already taken 2 line anti-TB drugs before getting pregnant, while 7/55 were diagnosed MDR TB and started therapy during pregnancy. After being counselled by clinicians, only 6/55 women chose abortion, while the rest decided to continue the pregnancy and undergo MDR therapy. Eleven women died (8 died during treatment and 3 died after completing treatment for unknown reasons). There was one stillbirth and one child died prematurely due to pneumonia. One woman and her child were lost to follow up. One woman had to terminate treatment due to hepatitis. Other cases were treated successfully. The results of the studies showed that, with an attentive follow-up and appropriate therapy, MDR-TB pregnant women can be cured and have a positive maternal outcome, and should therefore be given the option to continue with a pregnancy. The results also showed that a delay in, or default, MDR treatment were the main causes of mortality and morbidity for mothers and babies.
Follow-up actions for TB therapy include checking for a woman's drug consumption, clinical symptoms of anti-TB drug adverse effects, and liver function tests. Good compliance with TB treatment in pregnant women led to better maternal outcome and TB recovery rate. These studies showed that adequate health services and directly observed therapy (DOT) could greatly contribute to women's compliance, and, therefore, to treatment success.
There were 3 studies on follow-up of TB therapy for pregnant women and all were on INH prophylaxis. Pregnant women with LTBI were offered 6 months of prophylaxis with INH. The compliance rate was low, possibly due to the women's concern about hepatitis and other adverse effects, and the lack of referral services for treatment evaluation and action from health care providers to ensure compliance. Compared to the general population, pregnant women were less likely to initiate the INH prophylaxis – 52.1% vs 14.7%. In all 3 studies, the completion rate among pregnant women was very low (14.7%, 9.3%, and 21.2%). All studies recommended that health care providers implement better follow-up strategies to increase patient compliance in the prenatal and post-partum periods, ensure follow-up of drug adverse effects, and not dispense INH quantities covering more than 30 days of treatment at each visit.
Results
Summary of Included Studies
Thirty five studies were selected for detailed review. Good inter-rater agreement on the quality assessment of the studies was found (K = 0.70). There were 6 studies with good quality (++), 21 with medium quality (+), and 8 with low quality (−). The reasons to include studies with low quality is the limited number of studies in selected topics and one of the purpose of this review is to have an overall evaluation on the situation of research on TB care for pregnant women.
Table 1 shows the characteristics of the reviewed studies and the quality evaluation scores. Concerning the study types, there were 14 cohort studies, 8 before and after studies, 6 cross sectional studies, 4 case control studies, 2 controlled trials, and 1 qualitative study.
Concerning the topics addressed, some studies addressed multiple topics: 8 studies covered 2 topics and 1 study covered 3. The topic of TB diagnosis/screening was presented in 21 studies, 14 studies addressed TB treatment (including 4 on MDR treatment), 2 studies were on TB prevention, specifically on LTBI prophylaxis, and 3 studies were on the follow-up of women's compliance to the TB prophylaxis therapy. All studies were published after 1975, with a majority (26/35) published after 1999. Data originated from 16 different countries. Fifteen studies were carried out in resource-rich countries: US (11 studies), UK (3), and the Netherlands (1), while 20 were carried out in resource-limited countries: South Africa (6 studies), India (3), and the remaining countries had 1 each. One group of researchers in South Africa conducted 3 studies, other studies were all conducted by different groups of researchers. In all, 81093 people were enrolled in these studies, 37404 of whom in the study group (pregnant women with TB) and 43689 in the controlled/comparison groups (pregnant women without TB or/and non-pregnant women with TB). The large total was mostly due to one population based study involving 61016 people.
TB Diagnosis and Screening for Pregnant Women
The procedures for TB screening and diagnosis for pregnant women described in the reviewed studies include the tuberculin sensitivity test (TST/PPD) followed by the sputum test (Acid-fast bacillus - AFB) and the shielded chest X-ray. The AFB smear test appears to have low sensitivity in pregnant women, but is still used in low resource settings as part of the procedure for diagnosing active TB due to its low cost and simple technique. AFB culture was used as a confirmation of diagnosis, but is time consuming and not available in low resource settings. Another technique, fluorescence microscopy, was recommended as a substitute for AFB culture because it is cheaper. This procedure is for pulmonary TB cases and cannot identify extra pulmonary TB without additional tests and presence of physical TB symptoms. A clinical examination comprising a questionnaire tracking TB history and detecting TB clinical symptoms was also used and was proven to increase reliability of TB screening and diagnosis when combined with paraclinical tests. Some authors recommended not using chest X-ray for pregnant women if there were no clinical symptoms of TB.
TST is used widely as the first step in TB screening and diagnosis and to identify LTBI. Studies showed that pregnancy does not affect the sensitivity of this test, but its result can be affected by HIV infection or any situation that severely weakens the immune system (such as disseminated TB), as these could lead to false negative results. BCG vaccination can also lead to TST positive results in healthy women. In a high HIV prevalence setting, other tests and clinical symptoms should therefore be taken into account in diagnosing TB and the TST and anergy skin tests (the latter is used to evaluate whether the immune system is functioning properly or not and can indicate whether the results of the other skin test are reliable) are recommended as a TB screening method in the prenatal care procedures. In populations in which the majority of people are BCG vaccinated or their vaccination status is uncertain, TST is discouraged and IGRA is recommended for TB screening and diagnosis.
Concerning the IGRA test, one study in Kenya compared results of this test with the TST in screening for TB and showed the advantage of the IGRA test over TST in TB screening and diagnosis for HIV positive pregnant women, since its sensitivity is not affected by HIV infection. Two studies proved the value of IGRA in detecting LTBI in pregnant women, since the results of this test were not affected by BCG vaccination, thus avoiding the TST false positive result and the unnecessary, consequent INH prophylaxis.
Prevention
TB prevention includes BCG vaccination in childhood and INH prophylaxis for LTBI positive people. There were 2 studies on TB prevention and both were on INH prophylaxis for LTBI pregnant women. Both were conducted in the US with pregnant women of foreign origin and LTBI was diagnosed by TST. One study showed a low completion rate of INH therapy ( 9.3%) and the other showed a high risk of INH toxic hepatitis, with pregnant women having a 2.5 fold greater risk of INH hepatitis than non-pregnant women (but this result was not statistically significant due to the small number of women). The 2 studies found that the main reason for this discouraging result was a lack of follow up and referral services for pregnant women undergoing INH prophylaxis.
Treatment
There were 375 pregnant women with TB in the 14 studies on TB treatment. Treatment outcome was generally positive, with 332/375 women cured (confirmed by AFB culture conversion). In terms of mortality, 25 women died during treatment, 11 of whom died due to meningitis TB, 11 due to MDR-TB, 2 due to acute respiratory distress syndrome (ARDS), and 1 due to a non-TB related reason (massive pulmonary embolism). Other than mortality, the negative treatment outcomes included 4 treatment failures, 4 cases of residual functional deficit, 7 treatment terminations due to adverse drug effects, and 3 cases of treatment abandonment.
In terms of pregnancy outcome, only 11 women chose to terminate the pregnancy when they discovered their TB situation, while the others continued the pregnancy and underwent TB therapy. Among the pregnant women undergoing TB therapy, 332 women gave birth, 1 had a therapeutic abortion, 3 had miscarriages, 3 had stillborns, and 25 died. Concerning the 332 cases of mothers who gave birth, 4 infants died shortly after birth due to pneumonia and prematurity, 2 were HIV positive, 1 had active TB, 2 had LTBI, 50 were low birth weight, and 7 had growth restriction. Studies also showed that HIV infected women were more likely to choose pregnancy termination and had higher mortality and morbidity rates, even with intensive TB treatment combined with HIV treatment.
The first line drugs used for pregnant women in the studies included INH, ethambutol (ETB), rifampicin (RIF), and, in some cases of extra pulmonary TB, pyrazinamid (PZA). In the MDR cases, the second line drugs, including drugs of the amino glycosides group, fluoroquinolone, thioamides, cycloserines, and terizidone, were used in combination with effective first line drugs, and the treatment regime depended on the drug resistance situation of the individual cases.
Regarding effectiveness and safety of anti-TB drugs, results of the reviewed studies showed no significant association between child abnormality and mother's exposure to anti-TB drugs, both for 1 and 2 line anti-TB drugs during pregnancy. Other significant adverse effects were recorded in a very small number of pregnant women (2 cases of drug induced hepatitis, 2 of PZA allergy, 2 of sensorineural deafness, and 1 of severe nausea and jaundice) and led to termination of therapy without mortality. Streptomycin (SM) was not used in any studies because of its potential risk of deafness in babies. However, a retrospective study was conducted, checking the hearing capacity of 30 children whose mothers received SM injection during pregnancy, and found no significant effect, with only one case of deafness possibly linked to the mother's use of SM. Authors of this study recommend only using SM after the 2 trimester if really necessary.
There were 4 studies with 55 pregnant women on MDR treatment. Unlike other studies, in these TB was detected in all the women before pregnancy. More specifically, 48/55 of the women had been diagnosed with MDR TB and had already taken 2 line anti-TB drugs before getting pregnant, while 7/55 were diagnosed MDR TB and started therapy during pregnancy. After being counselled by clinicians, only 6/55 women chose abortion, while the rest decided to continue the pregnancy and undergo MDR therapy. Eleven women died (8 died during treatment and 3 died after completing treatment for unknown reasons). There was one stillbirth and one child died prematurely due to pneumonia. One woman and her child were lost to follow up. One woman had to terminate treatment due to hepatitis. Other cases were treated successfully. The results of the studies showed that, with an attentive follow-up and appropriate therapy, MDR-TB pregnant women can be cured and have a positive maternal outcome, and should therefore be given the option to continue with a pregnancy. The results also showed that a delay in, or default, MDR treatment were the main causes of mortality and morbidity for mothers and babies.
Follow-up
Follow-up actions for TB therapy include checking for a woman's drug consumption, clinical symptoms of anti-TB drug adverse effects, and liver function tests. Good compliance with TB treatment in pregnant women led to better maternal outcome and TB recovery rate. These studies showed that adequate health services and directly observed therapy (DOT) could greatly contribute to women's compliance, and, therefore, to treatment success.
There were 3 studies on follow-up of TB therapy for pregnant women and all were on INH prophylaxis. Pregnant women with LTBI were offered 6 months of prophylaxis with INH. The compliance rate was low, possibly due to the women's concern about hepatitis and other adverse effects, and the lack of referral services for treatment evaluation and action from health care providers to ensure compliance. Compared to the general population, pregnant women were less likely to initiate the INH prophylaxis – 52.1% vs 14.7%. In all 3 studies, the completion rate among pregnant women was very low (14.7%, 9.3%, and 21.2%). All studies recommended that health care providers implement better follow-up strategies to increase patient compliance in the prenatal and post-partum periods, ensure follow-up of drug adverse effects, and not dispense INH quantities covering more than 30 days of treatment at each visit.
