- Email: [email protected]
The prevalence of active tuberculosis infection among pregnant women is not increasing in the United States
Accepted Manuscript The prevalence of active tuberculosis infection among pregnant women is not increasing in the United States Jason L. Salemi, PhD, MPH, Hamisu M. Salihu, MD, PhD PII:
S0002-9378(17)30727-5
DOI:
10.1016/j.ajog.2017.05.064
Reference:
YMOB 11711
To appear in:
American Journal of Obstetrics and Gynecology
Received Date: 14 April 2017 Accepted Date: 31 May 2017
Please cite this article as: Salemi JL, Salihu HM, The prevalence of active tuberculosis infection among pregnant women is not increasing in the United States, American Journal of Obstetrics and Gynecology (2017), doi: 10.1016/j.ajog.2017.05.064. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Letters to the Editor
Title: The prevalence of active tuberculosis infection among pregnant women is not increasing in
Jason L. Salemi, PhD, MPHa; Hamisu M. Salihu, MD, PhDa
SC
a
RI PT
the United States
M AN U
Department of Family and Community Medicine, Baylor College of Medicine, Houston, Texas, USA
Corresponding author: Jason L. Salemi, PhD, MPH; Baylor College of Medicine, 3701 Kirby Dr., Suite 600 (MS: BCM700), Houston, TX 77098; (713) 798-4698 ([email protected])
Conflict of interest disclosure: The authors have no conflicts of interest to declare regarding the
465
AC C
EP
Word count:
TE D
publication of this letter to the editor.
ACCEPTED MANUSCRIPT
LETTER TO THE EDITOR A recent study published in AJOG by El-Messidi et al1 reported the prevalence of tuberculosis (TB) among pregnant women in the United States (US) to be 26.6 per 100,000. The authors also observed
RI PT
that the TB rate (per 100,000) among pregnant women increased, on average, from 19.2 in 2003 to 40.6 in 2011. If valid, the authors’ findings would be alarming: (1) a TB rate during pregnancy that is more than five times higher than that in the general US population;2 (2) an average 9.8% increase in the TB rate
SC
each year, in contradiction to consistent reports of decreasing TB rates in the US attributable to
improvements in TB control. We respectfully disagree with the findings published by El-Messidi and
M AN U
colleagues.
In the nationally-representative database used in the study, the Nationwide Inpatient Sample (NIS), TB and other clinical conditions are defined using ICD-9-CM codes. TB-affected pregnancies are those with one or more of the following codes: 010-018.x (TB); 137.x (sequelae of TB), 647.3 (TB complicating pregnancy/childbirth). However, the authors also included in their case definition V12.01
TE D
(personal history of TB), which in the absence of other TB codes is not indicative of active TB. Based on our re-analysis of 2002-2014 NIS data, approximately 75% of all TB cases in the authors’ study have only the V12.01 code. By excluding V12.01 from the case definition, the authors’ inexplicably high TB
EP
prevalence of 26.6 per 100,000 declines to 5.9, and the implausibly high temporal increases in TB rates during pregnancy disappear (Figure 1). This explains why the authors reported increases in the rates of
AC C
only nonpulmonary TB (whose definition includes V12.01), but not pulmonary TB. The authors support their increasing TB trends citing Albalak et al who found an increasing proportion of TB cases that were nonpulmonary; however, Albalak et al also reported significant decreases in TB and TB/HIV-coinfection rates from 1993-2004.3
El-Messidi et al also estimated associations (i.e., adjusted odds ratios [OR]) between TB in pregnancy and maternal/fetal complications. It is likely that their reported ORs underestimated true associations due to a TB case pool diluted by pregnant women with a history of TB but without active TB.4 This is evidenced by the substantially lower rates of sepsis, pneumonia, mechanical ventilation, and blood
ACCEPTED MANUSCRIPT
transfusion, as well as the lower mean length of stay (3.4 vs. 5.1 days) and lower direct medical cost ($8,000 vs. $12,000) among pregnant women with only the V12.01 code versus those with one or more of the aforementioned ‘active TB’ codes.
RI PT
We commend the authors on their study aims but feel their inclusion of the V12.01 code, which comprises most of their TB cases and has been excluded in other studies4, results in gross overestimation
AC C
EP
TE D
M AN U
SC
of both the prevalence and temporal trends in relevant TB infection among pregnant women in the US.
ACCEPTED MANUSCRIPT
REFERENCES 1.
El-Messidi A, Czuzoj-Shulman N, Spence AR, Abenhaim HA. Medical and obstetric outcomes among pregnant women with tuberculosis: a population-based study of 7.8 million births. Am J
2.
RI PT
Obstet Gynecol. 2016;215(6):797.e791-797.e796. Centers for Disease Control and Prevention. Tuberculosis Incidence in the United States, 19532015. 2016; https://www.cdc.gov/tb/statistics/tbcases.htm. Accessed April 13, 2017.
Albalak R, O'Brien RJ, Kammerer JS, et al. Trends in tuberculosis/human immunodeficiency
SC
3.
virus comorbidity, United States, 1993-2004. Archives of internal medicine. 2007;167(22):2443-
LaCourse SM, Greene SA, Dawson-Hahn EE, Hawes SE. Risk of Adverse Infant Outcomes Associated with Maternal Tuberculosis in a Low Burden Setting: A Population-Based Retrospective Cohort Study. Infectious diseases in obstetrics and gynecology.
EP
TE D
2016;2016:6413713.
AC C
4.
M AN U
2452.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 1. Temporal trends in the prevalence of tuberculosis in pregnancy (per 100,000 delivery-related hospitalizations), by tuberculosis case definition, Nationwide Inpatient Sample, 2002-2014
The X-axis represents the year of discharge and the Y-axis represents the percent of inpatient discharges with tuberculosis infection. Lines represent the trend estimated by joinpoint regression. Values represent the annual percent change (APC), point estimate (95% confidence interval). Tuberculosis case definitions, ICD-9-CM codes: (1) El-Messidi et al included 010.x – 018.x, 137.x, 647.3, V12.01; (2) V12.01 alone included V12.01; (3) corrected definition of ‘active TB’ included 010.x – 018.x, 137.x, 647.3
S0002-9378(17)30727-5
DOI:
10.1016/j.ajog.2017.05.064
Reference:
YMOB 11711
To appear in:
American Journal of Obstetrics and Gynecology
Received Date: 14 April 2017 Accepted Date: 31 May 2017
Please cite this article as: Salemi JL, Salihu HM, The prevalence of active tuberculosis infection among pregnant women is not increasing in the United States, American Journal of Obstetrics and Gynecology (2017), doi: 10.1016/j.ajog.2017.05.064. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Letters to the Editor
Title: The prevalence of active tuberculosis infection among pregnant women is not increasing in
Jason L. Salemi, PhD, MPHa; Hamisu M. Salihu, MD, PhDa
SC
a
RI PT
the United States
M AN U
Department of Family and Community Medicine, Baylor College of Medicine, Houston, Texas, USA
Corresponding author: Jason L. Salemi, PhD, MPH; Baylor College of Medicine, 3701 Kirby Dr., Suite 600 (MS: BCM700), Houston, TX 77098; (713) 798-4698 ([email protected])
Conflict of interest disclosure: The authors have no conflicts of interest to declare regarding the
465
AC C
EP
Word count:
TE D
publication of this letter to the editor.
ACCEPTED MANUSCRIPT
LETTER TO THE EDITOR A recent study published in AJOG by El-Messidi et al1 reported the prevalence of tuberculosis (TB) among pregnant women in the United States (US) to be 26.6 per 100,000. The authors also observed
RI PT
that the TB rate (per 100,000) among pregnant women increased, on average, from 19.2 in 2003 to 40.6 in 2011. If valid, the authors’ findings would be alarming: (1) a TB rate during pregnancy that is more than five times higher than that in the general US population;2 (2) an average 9.8% increase in the TB rate
SC
each year, in contradiction to consistent reports of decreasing TB rates in the US attributable to
improvements in TB control. We respectfully disagree with the findings published by El-Messidi and
M AN U
colleagues.
In the nationally-representative database used in the study, the Nationwide Inpatient Sample (NIS), TB and other clinical conditions are defined using ICD-9-CM codes. TB-affected pregnancies are those with one or more of the following codes: 010-018.x (TB); 137.x (sequelae of TB), 647.3 (TB complicating pregnancy/childbirth). However, the authors also included in their case definition V12.01
TE D
(personal history of TB), which in the absence of other TB codes is not indicative of active TB. Based on our re-analysis of 2002-2014 NIS data, approximately 75% of all TB cases in the authors’ study have only the V12.01 code. By excluding V12.01 from the case definition, the authors’ inexplicably high TB
EP
prevalence of 26.6 per 100,000 declines to 5.9, and the implausibly high temporal increases in TB rates during pregnancy disappear (Figure 1). This explains why the authors reported increases in the rates of
AC C
only nonpulmonary TB (whose definition includes V12.01), but not pulmonary TB. The authors support their increasing TB trends citing Albalak et al who found an increasing proportion of TB cases that were nonpulmonary; however, Albalak et al also reported significant decreases in TB and TB/HIV-coinfection rates from 1993-2004.3
El-Messidi et al also estimated associations (i.e., adjusted odds ratios [OR]) between TB in pregnancy and maternal/fetal complications. It is likely that their reported ORs underestimated true associations due to a TB case pool diluted by pregnant women with a history of TB but without active TB.4 This is evidenced by the substantially lower rates of sepsis, pneumonia, mechanical ventilation, and blood
ACCEPTED MANUSCRIPT
transfusion, as well as the lower mean length of stay (3.4 vs. 5.1 days) and lower direct medical cost ($8,000 vs. $12,000) among pregnant women with only the V12.01 code versus those with one or more of the aforementioned ‘active TB’ codes.
RI PT
We commend the authors on their study aims but feel their inclusion of the V12.01 code, which comprises most of their TB cases and has been excluded in other studies4, results in gross overestimation
AC C
EP
TE D
M AN U
SC
of both the prevalence and temporal trends in relevant TB infection among pregnant women in the US.
ACCEPTED MANUSCRIPT
REFERENCES 1.
El-Messidi A, Czuzoj-Shulman N, Spence AR, Abenhaim HA. Medical and obstetric outcomes among pregnant women with tuberculosis: a population-based study of 7.8 million births. Am J
2.
RI PT
Obstet Gynecol. 2016;215(6):797.e791-797.e796. Centers for Disease Control and Prevention. Tuberculosis Incidence in the United States, 19532015. 2016; https://www.cdc.gov/tb/statistics/tbcases.htm. Accessed April 13, 2017.
Albalak R, O'Brien RJ, Kammerer JS, et al. Trends in tuberculosis/human immunodeficiency
SC
3.
virus comorbidity, United States, 1993-2004. Archives of internal medicine. 2007;167(22):2443-
LaCourse SM, Greene SA, Dawson-Hahn EE, Hawes SE. Risk of Adverse Infant Outcomes Associated with Maternal Tuberculosis in a Low Burden Setting: A Population-Based Retrospective Cohort Study. Infectious diseases in obstetrics and gynecology.
EP
TE D
2016;2016:6413713.
AC C
4.
M AN U
2452.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figure 1. Temporal trends in the prevalence of tuberculosis in pregnancy (per 100,000 delivery-related hospitalizations), by tuberculosis case definition, Nationwide Inpatient Sample, 2002-2014
The X-axis represents the year of discharge and the Y-axis represents the percent of inpatient discharges with tuberculosis infection. Lines represent the trend estimated by joinpoint regression. Values represent the annual percent change (APC), point estimate (95% confidence interval). Tuberculosis case definitions, ICD-9-CM codes: (1) El-Messidi et al included 010.x – 018.x, 137.x, 647.3, V12.01; (2) V12.01 alone included V12.01; (3) corrected definition of ‘active TB’ included 010.x – 018.x, 137.x, 647.3