Clinical Infectious Diseases Advance Access published November 26, 2014
1 Complex anemia in tuberculosis: the need to consider causes and timing while
cr ipt
designing interventions
Peter A. Minchella1, Simon Donkor2, Olumuyiwa Owolabi2, Jayne S. Sutherland2, Joann M. McDermid1
2
Vaccinology Theme, Medical Research Council Unit, Fajara, The Gambia
us
Division of Nutritional Sciences, Cornell University, Ithaca, NY, USA
Corresponding Author: Joann M. McDermid, Division of Nutritional Sciences, 310
an
Savage Hall, Cornell University, Ithaca, NY 14853, Phone: +1-607-255-2490, Fax: +1607-255-1033, Email:
[email protected] M
Alternate Corresponding Author: Peter A. Minchella, Division of Nutritional Sciences, 308 Savage Hall, Cornell University, Ithaca, NY 14853, Phone: +1-765-714-5717, Fax:
pt ed
+1-607-255-1033
Summary of Main Findings (38 words): Anemia and tuberculosis comorbidity is common, and effective resolution of anemia must link interventions to causes. Inflammation anemia responds to tuberculosis treatment but iron-deficiency anemia
ce
does not. Monitoring iron-related biomarkers can inform effective and appropriately
Ac
timed anemia interventions.
© The Author 2014. Published by Oxford University Press on behalf of the Infectious Diseases Society of America. All rights reserved. For Permissions, please e‐mail:
[email protected].
Downloaded from http://cid.oxfordjournals.org/ at Monash University on December 4, 2014
1
2
Abstract
cr ipt
Background: Anemia is common in tuberculosis and multiple etiologies necessitate
targeted interventions. The proportion of iron-responsive anemia due to iron deficiency compared to iron-unresponsive anemia due to impaired iron absorption/redistribution from tuberculosis-associated immune activation or inflammation is unknown. This
us
Methods: Baseline hemoglobin, ferritin, hepcidin, soluble transferrin receptor (sTfR)
an
and transferrin were measured in 45 confirmed pulmonary tuberculosis cases, 47 tuberculin skin test (TST)-positive controls and 39 TST-negative controls in The Gambia.
M
TB cases were additionally followed two and six months after tuberculosis treatment initiation. Mutually exclusive anemia categories based on iron biomarker concentrations were iron-deficiency anemia (IDA), anemia of inflammation (AI) and multifactorial
pt ed
anemia (IDA+AI).
Results: Anemia was more frequent in tuberculosis cases (67%) compared to TSTpositive (36%) or TST-negative (21%) controls. AI was the predominant anemia at tuberculosis diagnosis, declining from 36% to 8% after six months of treatment;
ce
however, a corresponding reduction was not evident for anemia with iron-responsive components (IDA, IDA+AI). Iron biomarkers discriminated between active tuberculosis
Ac
and TST-positive or TST-negative controls, as well as between active untreated and treated tuberculosis. This was most noticeable for hepcidin, which decreased from a median of 84.0 ng/ml at diagnosis to 9.7 ng/ml after two months (p10mm was
cr ipt
considered TST-positive (n=47) and ≤10mm TST-negative (n=39) due to Bacille-
Calmette Guerin (BCG) vaccination and high prevalence of environmental mycobacteria in this setting. Sputum samples were stained with auramine and the Ziehl-Nielsen
method. Sputum was decontaminated using N-acetyl cysteine (NALC)-NaOH before
us
slopes. Positive cultures were confirmed by Ziehl-Nielsen smear.
an
All TB cases underwent a clinical assessment at recruitment (defined as TB diagnosis) and at two and six months following TB treatment initiation. The clinical
M
assessment included determination of disease severity by chest x-ray, comorbidity assessment and TB-specific clinical questions such as the presence of side pain, cough duration and weight loss. Malaria screening was conducted using a rapid diagnostic
pt ed
test, and those with malaria were excluded from participation. Blood samples were obtained from TB cases at all three time points and at baseline only for TST-positive and TST-negative groups with storage at -70oC until analysis. At the time these data were collected, TB treatment success rate in The Gambia was >85% [26] and national
ce
TB treatment guidelines included an intensive phase of isoniazid, rifampicin, ethambutol, and pyrazinamide for two months followed by isoniazid and rifampicin for four months
Ac
[27].
Downloaded from http://cid.oxfordjournals.org/ at Monash University on December 4, 2014
culturing in Bactec vials (Becton Dickinson, USA) and on paired Lowenstein-Jensen
7 Demographic, Anthropometric and Clinical Characteristics Data was collected on participant age, gender, body mass index (BMI), mid-
(WBC) and mean corpuscular volume (MCV). Iron Status Biomarkers and Anemia Classification
cr ipt
upper arm circumference (MUAC), HIV-status, hemoglobin (Hb), white blood cell count
Plasma or serum ferritin (Immuno-biological Laboratories, Germany), soluble
us
Technologies, USA) were measured by ELISA. Hepcidin was measured using a
an
competitive enzyme immunoassay (Bachem, USA). All biomarker assays were optimized for use with plasma and serum and a sensitivity analysis confirmed specimen
M
type did not affect interpretation. Test samples, standards, and controls were assayed in duplicate and concentrations interpolated from 4-parameter logistic standard curves (log/log curves for sTfR). All standard curves were generated using SoftmaxPro 6
pt ed
(Molecular Devices, USA). Samples with an intra-assay coefficient of variation >15%, were re-assayed. Lower limits of detection (LoDs) for all assays, with the exception of hepcidin, were defined by the manufacturer. For hepcidin, the LoD (0.2 ng/ml) was interpolated at three standard deviations from the all plate mean (e.g. wells that
ce
contained diluent in lieu of hepcidin standard or sample). Anemia was defined according to World Health Organization criteria for males Hb