Pregnancy increases both a woman's risk of developing tuberculosis and her likelihood to seek health care. A symptom screen that is effective during pregnancy would therefore be valuable. We assessed a World Health Organization (WHO)-recommended TB symptom screen in an antenatal setting in India.
This was a cross-sectional study of women presenting to the antenatal clinic at a government hospital in Pune, India. We collected medical history and administered the WHO TB symptom screen: cough of any duration, fever, weight loss, or night sweats. All HIV positive women, and HIV negative women with at least one symptom, provided sputum samples, which were analyzed by direct sputum smear and GeneXpert.
From November 2014 to May 2015, 479 pregnant women enrolled. Median gestational age was 29 (IQR: 22-34) weeks. Of the 34 (7 %) HIV positive women, 33 (97 %) were on antiretroviral therapy with a median CD4 count of 547 (IQR: 403-707).
The symptom screen was positive in 112 (23%) women (13 HIV positive, 99 HIV negative.) Multivariate analysis showed that a positive symptom screen was significantly associated with living in a rural area (OR 2.3, CI: 1.2-4.3, p=0.009) and having HIV (OR 2.2, CI: 1.0-4.5, p=0.04)
Only 67 (50%) women with a positive symptom screen or HIV (Figure 1) were able to provide sputum samples. Multivariate analysis showed that fever (OR 3.9, CI: 1.5-9.9, p=0.004) and current cough (OR 2.4, CI: 1.0-5.5, p=0.041) were significantly associated with ability to provide a sputum sample. GeneXpert could not analyze 13 (19%) samples due to low volume. Therefore, of the 133 asked to provide sputum samples, only 54 received GeneXpert analysis. The screening process, including sputum collection took a median of 14 minutes (IQR 10-16). No active TB cases were diagnosed.
The WHO TB symptom screen did not improve TB diagnosis in pregnancy. Many women with positive screens were unable to produce enough sputum for GeneXpert analysis. These findings suggest that only patients who report fever or current cough should be asked to provide sputum samples. Given the time and expense of implementing this screen, alternate strategies should be considered for identifying active TB in pregnant women.
R. Bhosale, None
V. Mave, None
J. Elf, None
N. Pradhan, None
N. Nevrekar, None
S. Patil, None
A. Kagal, None
S. Joshi, None
A. Chandanwale, None
A. Gupta, None
J. S. Mathad, None