Kebede and Mamo BMC Pulmonary Medicine (2024) 24:527 BMC Pulmonary Medicine

https://doi.org/10.1186/s12890-024-03350-w

RESEARCH Open Access

Multidrug‑resistant tuberculosis treatment

outcomes and associated factors at Yirgalem
General Hospital, Sidama Region, South
Ethiopia: a retrospective cohort study
Assefa Hamato Kebede1,2 and Hassen Mamo2*

Abstract
Background The spread of multidrug-resistant tuberculosis (MDR-TB) poses a significant challenge to TB control
efforts. This study evaluated the treatment outcomes and associated factors among patients receiving treatment
for MDR-TB in southern Ethiopia.
Methods A retrospective follow-up study covering ten years, from 2014 to 2023, analyzed the records of confirmed
cases of pulmonary TB admitted to Yirgalem General Hospital, an MDR-TB treatment initiation center in the Sidama
Region. To compare the successful treatment outcomes across the years, a chi-square test of independence was con-
ducted. Bivariate and multivariable logistic regression models were used to identify factors associated with treatment
outcomes for MDR-TB.
Results Out of 276 confirmed MDR-TB cases, 4(1.4%) were diagnosed with resistance to second-line drugs (SLDs).
Overall, 138 patients achieved favourable treatment outcomes, resulting in a treatment success rate of 50.0% [95%
CI 44.1–55.9%]. Among these 138 patients, 105(76.1%, 95 CI 68.7–83.5%) were cured, while 33(23.9%, 95 CI 16.5–
31.3%) completed their treatment. The successful treatment outcomes varied significantly across the years, ranging
from 3.6% in 2020 to 90% in 2021. The analysis indicated a statistically significant difference in treatment outcomes
when considering data from 2014 to 2023 (χ2 = 44.539, p = 0.001). The proportion of patients with deaths, lost-to-
follow-up (LTFU), treatment failures and not evaluated were 7.9% [95% CI 4.8–11.2%], 10.9% [95% CI 7.2–14.6%), 2.2%
[95% CI 1.1–3.3%), and 28.9% [95% CI 23.7–34.2%] respectively. Individuals with a positive HIV status had signifi-
cantly lower odds of a favorable treatment outcome [AOR = 0.628, 95% CI (0.479–0.824), p = 0.018]. Similarly, patients
with a BMI of less than 18 are more likely to have unfavorable treatment outcomes compared to those with a BMI
of 18 or higher [AOR = 2.353, 95% CI 1.404–3.942, p < 0.001].
Conclusion The study revealed a concerning 1.4% prevalence of additional resistance to SLDs. The 50% rate of unfa-
vorable treatment among MDR-TB cases exceeds the target set by the WHO. A significant number of patients (10.9%)
were LTFU, and the 28.9% categorized as ‘not evaluated’ is also concerning. Enhanced strategic interventions are
needed to reduce such cases, and factors associated with poor treatment outcomes should receive greater attention.
Future prospective studies can further explore the factors influencing improved treatment success.

*Correspondence:
Hassen Mamo
binmamo@yahoo.com
Full list of author information is available at the end of the article

© The Author(s) 2024. Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0
International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long
as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if
you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or
parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated
otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not
permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To
view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
Kebede and Mamo BMC Pulmonary Medicine (2024) 24:527 Page 2 of 10

Keywords Multidrug-resistant tuberculosis, Treatment initiation center, Treatment outcome, Sidama

Introduction While significant progress has been made in TB surveil-

Despite remarkable progress in the control of tuberculo- lance, some important gaps remain. Some of a few areas
sis (TB) globally, the emergence of multidrug-resistant TB where knowledge gaps exist in TB surveillance include
(MDR-TB) and extensively drug-resistant TB (XDR-TB) under-reporting and case detection, DR-TB surveillance,
Mycobacterium tuberculosis strains particularly poses a pediatric TB surveillance, timeliness of data, sub-national
major challenge to effective treatment [1]. These strains and marginalized populations, and integration of data
are resistant to the most commonly used anti-TB drugs, sources [23]. TB is known to be under-reported, particu-
requiring longer and more complex treatment regimens larly in low-resource settings, and many cases go undi-
with more expensive and toxic drugs [2]. Moreover, agnosed or unreported due to limitations in healthcare
access to accurate and timely diagnosis of TB is crucial access, inadequate diagnostic capacity, and weak surveil-
for effective treatment. However, many regions, espe- lance systems [24, 25]. This study aimed at assessing drug
cially low- and middle-income countries, face challenges resistance, treatment outcomes, and associated factors in
in accessing quality diagnostic tools, such as GeneXpert a cohort of pulmonary MDR-TB patients enrolled in the
machines or culture-based methods [3–8]. This leads to programmatic management of DR-TB at Yirgalem Gen-
delays in diagnosis and initiation of treatment as well as eral Hospital (YGH), which is MDR-TB treatment initia-
evaluation of treatment outcomes. TB treatment typically tion center in Sidama National Regional State (SNRS),
involves a long course of multiple drugs taken over several southern Ethiopia.
months [9]. Ensuring patient adherence to the treatment
regimen is critical for successful outcomes and preventing
the development of drug resistance. However, adherence Methods
remains a further challenge due to various factors, includ- Study site
ing the complexity of the treatment, side effects of medi- A retrospective follow-up study was conducted at YGH
cations, and socioeconomic barriers [10]. in Yirgalem town, located in south Ethiopia, from Febru-
Co-infections and comorbidities are yet other chal- ary to December 2023. Yirgalem is situated 47 km away
lenges. TB often occurs alongside other health condi- from Hawassa city which serves as the capital of the
tions, such as HIV/AIDS, diabetes, and malnutrition and SNRS and is 319 km from Addis Ababa. The Hospital
complicate TB treatment, increase the risk of treatment provides inpatient and outpatient care to a population
failure, and lead to poor outcomes [11–14]. Integrated of over three million patients and also serves as a refer-
management of these conditions is essential but can be ral center for nearby areas such as Oromia and the South
challenging due to resource limitations and fragmented Ethiopia Peoples’ Regions administrative zones.
healthcare systems and demographic challenges among SNRS is one of the largest, recently emerged and most
others. Diagnosing and treating pediatric TB is more densely populated administrative Regions in the country.
challenging in that children often have paucibacillary According to the census and population projection con-
TB making it harder to detect through standard diag- ducted by the Ethiopia Statistical Service [26], the Region
nostic methods, pediatric-specific formulations of anti- has an estimated population of over 5 million. The major-
TB drugs are often limited, leading to the need for dose ity of the Region’s inhabitants are rural dwellers, and agri-
adjustments and off-label use; and children may face culture is the primary livelihood, similar to other Regions
more challenges with adherence, leading to sub-optimal in Ethiopia. The Region has over 600 public health facili-
treatment outcomes [15–18]. Furthermore, social and ties and 84 private clinics that are actively involved in TB
structural factors, such as poverty, stigma, lack of aware- treatments through the directly observed therapy (DOT)
ness, and weak healthcare systems, hinder early diag- course initiative. Private sector health facilities contrib-
nosis, treatment initiation, and successful treatment ute to TB control and management activity in Ethiopia
completion [19, 20]. [27, 28]. The Hospital is administratively governed by the
Disease surveillance is key to control and manage SNRS and has been serving as the only treatment initia-
infectious diseases in general [21]. Regular surveillance of tion center for MDR-TB since the end of 2013. YGH has
TB plays a crucial role in addressing drug resistance by been functioning as one of the country’s treatment initia-
providing valuable insights into the patterns and dynam- tion centers (TICs) for the past decade, providing medi-
ics of resistance development, treatment outcomes, and cal care for curative and rehabilitative treatments across
the effectiveness of public health interventions [22]. various units, including MDR-TB treatment services.
Kebede and Mamo BMC Pulmonary Medicine (2024) 24:527 Page 3 of 10

Additionally, the center is responsible for monitoring or the mycobacterial growth indicator tube (MGIT) sys-
several treatment follow-up centers (TFCs) for patients tem, which can take several weeks to obtain results [35].
undergoing ambulatory outpatient treatment care Chest X-rays and, in some cases, chest computed tomog-
through the DOT program. raphy (CT) scans are used to assess the extent and sever-
ity of pulmonary involvement, which can guide treatment
Study population and data collection decisions [36]. Due to the high co-occurrence of TB and
From January 2014 up to December 2023, a total of 308 HIV in Ethiopia, all TB patients, including those with
cases of MDR-TB were enrolled at the YGH MDR-TB MDR-TB, are tested for HIV [37]. HIV infection can
TIC. For this study, we included 276 cases of pulmonary impact the clinical presentation, treatment response, and
MDR-TB from the registry and medical records. Struc- management of MDR-TB. Finally, comprehensive clini-
tured data collection tools were developed based on the cal, social, and psychosocial assessments are conducted
national MDR-TB registration book and existing litera- to evaluate the patient’s overall health status, comorbidi-
ture to gather demographic and medical information. ties, and potential barriers to treatment adherence. This
Two trained nurses were employed to collect the data information is used to develop a personalized treatment
under supervision. The collected characteristics included plan and to identify any necessary support or interven-
sex, age, residence, year of enrollment, baseline informa- tions to ensure successful treatment outcomes.
tion such as sputum smear, treatment category (new or
re-treatment), prior drug exposure, HIV status, other Operational definitions
co-morbid diseases, confirmed method of diagnosis, All terms used in the study are operationally defined and
resistance type, body-mass-index (BMI), and treatment classified according to recommended guidelines [38]. Pre-
outcomes. sumptive TB refers to a patient who presents with symp-
toms or signs suggestive of TB (previously known as a
TB suspect). Rifampicin-resistant TB was defined as the
MDR‑TB diagnosis in Ethiopia detection of resistance to RIF using either phenotypic or
The diagnosis of MDR-TB in Ethiopia, as it is elsewhere, genotypic methods, with or without co-resistance to other
involves a multi-step process [29–32]. The initial step is anti-TB drugs. Multidrug-drug resistant TB (MDR-TB)
a clinical diagnosis. Patients with a history of previous was defined as caused by MTB strains that are resistant
TB treatment, especially those who have relapsed, failed to at least both and RIF and INH. The term cured was
treatment, or defaulted, are at an increased risk of devel- used to indicate patients with no evidence of treatment
oping MDR-TB and are considered presumptive cases of failure, where three or more consecutive cultures taken
the condition. Additionally, patients presenting with per- at least 30 days apart showed negative results after the
sistent symptoms of TB, such as cough, fever, weight loss, intensive phase. Treatment completed referred to patients
and night sweats, despite receiving appropriate first-line who showed no evidence of failure but did not have a
TB treatment, are also considered presumptive cases of record of three or more consecutive cultures taken at least
MDR-TB. 30 days apart showing negative results after the intensive
The presumptive TB cases then undergo bacteriological phase. Treatment failure was defined as a patient whose
confirmation, during which sputum samples are collected treatment was terminated or required a permanent regi-
and examined microscopically for the presence of acid- men change of at least two anti-TB drugs. The term died
fast bacilli (AFB). If AFB are detected, the samples are referred to a confirmed TB patients who passed away for
further tested using rapid molecular diagnostics, such as any reason before or during the course of treatment. Lost-
the Xpert MTB/RIF assay [33], to detect the presence of to-follow-up (LTFU) included patients whose treatment
MTB and determine resistance to RIF, a key first-line TB was interrupted for two consecutive months or more. Not
drug. Samples with RIF resistance are considered highly evaluated referred to a TB patient for whom no treatment
suggestive of MDR-TB and require further testing. outcome is assigned. This includes cases “transferred out”
Confirmed MDR-TB cases undergo comprehensive to another treatment unit as well as cases for whom the
drug susceptibility testing (DST) to determine resistance treatment outcome is unknown to the reporting unit.
patterns to a wider range of anti-TB drugs, including Treatment success or favorable treatment outcome was
INH, ethambutol, and injectable agents (e.g., kanamycin, used to describe patients who met the definitions of cure
amikacin, capreomycin) [34]. This information is cru- or treatment completion. Poor or unfavorable treatment
cial for designing an appropriate and effective treatment outcome was used to describe patients who died, were
regimen for the patient. DST is typically performed using LTFU, and/or experienced treatment failure during the
culture-based methods, such as the proportion method course of treatment.
Kebede and Mamo BMC Pulmonary Medicine (2024) 24:527 Page 4 of 10

Data quality control Table 1 Socio-demographic characteristics of MDR-TB patients

Data quality was ensured by comparing the collected at Yirgalem General Hospital MDR-TB treatment initiation center,
data with the structured checklist encoding. Checkups Sidama Region, South Ethiopia, 2014–2023, N = 276
were made in consultation with the designated nurses, Variable Frequency Percent
and regular communication was maintained with them.
The data was manually entered into Microsoft Excel Sex
spreadsheets after cleaning and exported to SPSS version Male 175 63.4
16 (SPSS Inc.) for analysis. Female 101 36.4
Age (year)
1–14 19 1.8
Data processing and analysis
15–25 116 42
Descriptive statistics were used to calculate frequencies
26–45 112 40.6
and percentages for categorical variables, while continu-
46–55 16 5.8
ous variables were summarized using mean and median
≥ 56 13 4.7
ranges. A comparison between male and female patients
Residence
was conducted to examine demographics, clinical factors,
and treatment outcomes, which were presented as per- Hawassa city 30 10.9

centage proportions. Descriptive and regression models Other (district towns) 246 89.1
were utilized to identify factors associated with favora- Occupation
ble and unfavourable treatment outcomes. To determine Govermental employees 96 34.8
factors independently associated with unfavorable treat- Non-employees 180 65.2
ment outcomes, stepwise backward logistic regression Enrollment year
analysis was employed. A statistical significance level of 2014 33 12
p < 0.05 was deemed to be significant. 2015 10 3.6
2016 18 6.5
2017 30 10.9
Results
2018 50 18.1
Socio‑demographic characteristics
2019 28 10.1
Between 2014 and 2023, a total of 308 cases were enrolled
2020 28 10.1
in the MDR-TB center at YGH for treatment initiation.
2021 10 3.6
However, 32 of these were excluded as they were referred
2022 33 12
from outside the SNRS. Therefore, only 276 cases were
2023 36 13
considered. Among these 276 cases, 175(63.4%) were
male patients and 101(36.4%) were female patients,
resulting in a male-to-female ratio of approximately 2:1.
The mean age was 28.95 ± 12.882 years, with a median age Additionally, 16(5.8%) patients had comorbid diseases
of 26 (interquartile range [IQR] = 20–35) years. Among other than HIV co-infections, with diabetes mellitus
the cases, 160(57.97%) patients were between 20 and being the most common. The average BMI was 17.1736,
35 years of age, with the minimum and maximum ages with a median IQR of 17.000 (15.2000–18.9000), and
being 1 and 80 years, respectively. The majority (89.1%) 79.3% of the patients had a baseline BMI value ≥ 18 kg/
of the patients were from rural areas of the region, and m2 (Table 2). All bacteriologically confirmed MDR-TB
the highest number of cases were from Aroresa, Chire, cases were also screened and treated based on confirma-
Arbegona, and Hula districts, which are located more tory drug susceptibility test (DST) results before and after
than 100 km away from the treatment initiation center, treatment initiated. Based on DST result, 14(5.1%) were
as well as Hawassa city. Moreover, 180(65.2%) of the par- resistant to both INH and RIF (13 retreatment and 1 new
ticipants were non-employees. In terms of the year of cases). Four (1.5%) MDR-TB cases were diagnosed with
treatment initiation, 50(18.1%) patients were enrolled in additional resistance to SLDs at enrollment, with levo-
2018, representing the highest proportion, followed by floxacin being the dominant drug resistance observed in
36(13.0%) patients enrolled in 2023 (Table 1). 3 MDR-TB female patients who had previous exposure to
ant-TB drugs.
Medical and anthropometric characteristics
Among the 276 cases, 210(76.1%) had previously received Treatment outcomes
anti-TB drugs (retreatment cases) the rest were new at The final treatment outcomes of 80 patients (29.1%,
enrollment. At enrollment, all patients were tested for 95% CI 23.74–34.24%) had no recorded treatment out-
HIV co-infection, and 21(7.6%) patients tested positive. comes in the treatment center. This includes 1 patient in
Kebede and Mamo BMC Pulmonary Medicine (2024) 24:527 Page 5 of 10

Table 2 Clinical and anthropometric characteristics of MDR-TB (cured + complete) among male participants was approx-
patients at Yirgalem General Hospital MDR-TB treatment initiation imately 1.6(84/54) times compared to females. Treatment
center, Sidama Region, South Ethiopia, 2014–2023, N = 276 outcomes varied significantly from year to year, with
Variable History of previous drug Overall, no(%) the lowest success rates observed in 2020 (3.6%), 2023
use, no(%) (11.1%), and 2022 (36.4%). Slight changes were noted
Yes (N = 210) No (N = 66) between 2016, 2017, and 2018, while the highest success
rate (90%) was recorded in 2021. The difference in suc-
HIV status cessful treatment outcomes between the years were sta-
Reactive 13(61.9) 8(38.1) 21(7.6) tistically significant (χ2 = 44.539, p = 0.001).
Non-reactive 197(77.3) 58(22.7) 255(92.4)
Co-morbid disease other than HIV
Yes 16(100) 0(0) 16(5.8) Factors affecting treatment outcomes
No 194(74.6) 66(25.4) 260(94.2) Out of variables tested in the bivariate and multivariable
Resistance profile logistic regression (Table 4), individuals with a positive
RIF only 197(76.4) 61(23.6) 258(93.5) HIV status had significantly lower odds of a favorable
RIF + INH 13(92.9) 1(7.1) 14(5.1) treatment outcome compared to those with a negative
Any resistance to SLD 3(75) 1(25) 4(1.4) HIV status (adjusted odds ratio (AOR) = 0.628, 95% CI
(Fluoroquinolone) 0.479–0.824, p = 0.018]. Individuals having a BMI < 18
Diagnostic method had significantly higher risk of having unfavourable treat-
GeneXpert 187(75) 60(33.1) 247(89.4) ment outcome compared to those having BMI of ≥ 18
Line probe Assay 14(82.4) 3(16.6) 17(6.1) (AOR = 2.353, 95% CI 1.404–3.942, p < 0.001). The AOR
Confirmed culture 5(55.6) 4(44.4) 9(3.3) of 0.628 suggests that being HIV positive is linked to a
Other (clinical) 3(100) 0(0) 3(1.1) higher risk of unfavorable outcomes compared to being
Baseline sputum smear test result HIV negative. The lower odds ratio indicates a significant
Positive 123(69.5) 54(30.5) 177(64.1) negative impact on treatment success among those who
Negative 85(89.5) 10(10.5) 95(34.9) are HIV reactive. The AOR of 2.353 indicates that having
Not done 2(50) 2(50) 4(1.4) a low BMI is associated with a significantly higher risk
BMI (kg/m2) of unfavorable outcomes. This means that underweight
< 18 37(64.9) 20(35.1) 57(20.7) patients are more likely to experience negative treatment
≥ 18 173(79) 46(21) 219(79.3) results. Overall, the analysis identifies HIV positivity and
Have drug adverse 120(43.5) low BMI as significant factors negatively impacting treat-
effects ment outcomes for MDR-TB patients.
BMI Body Mass Index, INH Isoniazid, RIF Rifampicin

Discussion
2017 (N = 30), 3 patients in 2018 (N = 50), 3 patients in The management of MDR-TB necessitates increased
2019 (N = 28), 26 patients in 2020 (N = 28), 18 patients resources for detection, successful treatment, and effec-
in 2022 (N = 33), and 29 patients in 2023 (N = 36). The tive reduction of its burden. In 2015, Ethiopia successfully
overall treatment success rate was 50%(138/276) [95% achieved the millennium development goals established
CI 44.1–55.9%]. Of these 138 favorable treatment cases, for TB and expressed its commitment to expedite the
105(76.1%, 95% CI 68.7–83.5%) patients were declared eradication of TB by 2035 as part of a national strategic
cured, while 33(23.9, 95% CI 16.5–31.3%) completed plan [39]. This plan emphasizes the implementation of
their treatment. In the ten-year retrospective follow-up robust TB case-finding strategies and the utilization of
study, 22 patients (7.9%, 95% CI 4.8–11.2%) died during rapid diagnostic technologies to address the gap in iden-
treatment, 30(10.9%, 95% CI 7.2–14.6%) were LTFU, 6 tifying missed TB cases and combat MDR-TB threat.
patients (2.2%, 95% CI 1.1–3.3%) experienced treatment Our study indicated that 76.1% of MDR-TB patients
failure, and 80 (28.9%, 95% CI 23.7–34.2%) fell into the had previously undergone treatment for TB, and 5.1%
‘not evaluated’ category (Table 3). exhibited resistance to both INH and RIF (for new/pre-
Regarding the adverse effects of the drugs, 120 out of vious exposure), while 1.4% of MDR-TB cases demon-
276(43.5%) developed drug side effects following the strated additional resistance to SLDs (pre-XDR/TB) at
initiation of treatment. The proportion of male patients the baseline and treated within the cohort of MDR-TB. A
with favourable treatment outcomes was 48.0% (84/175) study that analyzed an aggregated and individual patient
and that of female participants in the cohort was 53.5% data [40] using six eligible studies reporting treatment
54/101. However, the relative ratio of treatment success outcome on 1993 MDR-TB patients in Ethiopia found a
Table 3 Trends of treatment outcomes of MDR-TB patients at Yirgalem General Hospital MDR-TB treatment initiation center, Sidama Region, South Ethiopia, 2014–2023 (N = 276)
Enrollment year
Treatment outcome 2014 N = 33 2015 N = 10 2016 N = 18 2017 N = 30 2018 N = 50 2019 N = 28 2020 N = 28 2021 N = 10 2022 N = 33 2023 N = 36 Total
N = 276
Kebede and Mamo BMC Pulmonary Medicine

Favorable, no(%) 27(81.8) 7(70.0) 11(61.1) 18(60.0) 30(60.0) 19(67.9) 1(3.6) 9(90.0) 12(36.4) 4(11.1) 138(50.0)
Cured 21 6 10 12 20 16 1 8 7 4 105(76.1)
Female 7 2 2 3 7 7 0 6 3 1 38(36.2)
Male 14 4 8 9 13 9 1 2 4 3 67(63.8)
(2024) 24:527

Complete 6 1 1 6 10 3 0 1 5 0 33(23.9)
Female 2 1 0 3 5 0 0 1 4 0 16(48.5)
Male 4 0 1 3 5 3 0 0 1 0 17(51.5)
Unfavorable, no(%) 6(18.2) 3(30.0) 7(38.9) 12(40.0) 20(40.0) 9(32.1) 27(96.4) 1(10.0) 21(63.6) 32(88.9) 138(50.0)
Died 3 1 4 1 5 3 1 1 2 1 22(7.9)
Female 1 0 1 0 3 0 0 0 0 0 5(1.8)
Male 2 1 3 1 2 3 1 1 2 1 17(6.2)
LTFU 2 1 2 8 12 2 0 0 1 2 30(10.9)
Female 0 1 0 3 1 0 0 0 1 1 7(2.5)
Male 2 0 2 5 11 2 0 0 0 1 23(8.3)
Failure 1 1 1 2 0 1 0 0 0 0 6(2.2)
Female 0 0 1 2 0 1 0 0 0 0 4(1.4)
Male 1 1 0 0 0 0 0 0 0 0 2(0.7)
Not evaluated 0 0 0 1 3 3 26 0 18 29 80(28.9)
Female 0 0 0 0 3 1 8 0 10 9 31(38.8)
Male 0 0 0 1 0 2 18 0 8 20 49(61.2)
Page 6 of 10
Kebede and Mamo BMC Pulmonary Medicine (2024) 24:527 Page 7 of 10

Table 4 Bivariate and multivariable analysis of factors associated with treatment outcomes of MDR-TB treatment at YGH MDR-TB
treatment initiation center, Sidama Region, South Ethiopia, 2014–2023 (N = 276)
Variable Treatment outcome COR (95% CI) AOR (95% CI) p-value
Unfavorable no(%) Favorable no(%)

Age in years
< 24 56(53.3) 49(46.7) 1.000 1.000 0.132
24–44 68(51.1) 65(48.9) 0.510 (0.238–1.094) -
≥ 45 14(36.8) 24(63.2) 0.558(0.266–1.171)
HIV status
Non-reactive 122(47.8) 133(52.2) 1.000 1.000
Reactive 16(76.2) 5(23.8) 0.287 (0.2–0.70) 0.628 (0.479—0.824) 0.018
Non-HIV comorbidity
No 134(51.5) 126(48.5) 1.000 1.000
Yes 4(25.0) 12(75.5) 3.190 (1.003–10.152) 1.063(1.003–1.128) 0.077
History of TB drug use
No 36(54.5) 40(45.5) 1.000 1.000 0.257
Yes 102(48.6) 108(51.4) 0.787(0.452–1.371) 0.944(0.827–1.078)
BMI
≥ 18 98(44.7) 121(55.3) 1.000 1.000
< 18 40(70.2) 17(29.8) 2.905(1.552–5.438) 2.353(1.404–3.942) < 0.001

pooled treatment success of 59.2% highlighting treatment factor for MDR-TB. However, it is worth noting that pos-
success among MDR-TB patient was below acceptable sible mismanagement of anti-TB drug-sensitive cases at
range. In a study in the northwestern part of Ethiopia various levels can be a contributing factor to the emer-
which analyzed retrospective data from 2011 to 2021 gence of drug resistance. The finding of 1.4% of cases
[41], the treatment success rate was 77.1%. diagnosed with additional resistance to core SLDs in
A systematic review published in 2014 [42] reported our study is alarming and poses a threat to the national
high prevalence of MDR-TB in the range of 3.3–46.3%, TB end strategy, as the majority of cases remain undiag-
and two studies reported XDR-TB in the range of 1–4.4% nosed, putting the community at risk and increasing the
in Ethiopia. The review concluded that ‘most power- likelihood of disease progression towards another form
ful’ predictor of the emergence of MDR-TB reported in of MDR-TB, such as XDR-TB. Notably, the findings from
Ethiopia was previous exposure to anti-TB drug treat- our study recommend future community-based studies
ment. This review indicated that MDR-TB in Ethiopia is a to obtain updated a nationwide figure.
serious public health problem that needs to be addressed In the current cohort study, 50.0% of patients
urgently. A subsequent similar systematic review and achieved successful treatment outcomes is lower than
meta-analysis study published in 2017 [43] reported the rates reported in hospital-based studies conducted
an overall prevalence of MDR-TB among newly diag- in northwest and northeast Ethiopia [41]. Our study
nosed and previously treated TB patients to be 2% (95% notably reports an 7.9% [95% CI 4.8–11.2%] death rate
CI 1–2%) and 15% (95% CI 12–17%), respectively for among patients during the treatment course, which is
16 eligible studies assessed. This study also found that comparable with a hospital-based study by Belachew
previously treated TB patients were at a higher risk of et al. 9.3% [41]. Furthermore, all patients diagnosed
developing a MDR-MTB infection compared to newly with additional resistance to SLDs died after enroll-
diagnosed cases. The study concluded that the burden of ment, which may reflect the level of patient-centered
MDR-TB remains high in Ethiopian settings, especially in services in the treatment center. The fact that all
previously treated TB cases although for the past 10 years patients with additional resistance died highlights the
(2006 to 2014) the overall MDR-TB prevalence showed a need for improved patient-centered services and treat-
stable time trend. ment outcomes. Furthermore, our study found that
Later studies conducted in the eastern [44], central 50.0% of MDR-TB cases had unsuccessful treatment
[45], and southern Ethiopia [46, 47] have implicated pre- outcomes in the entire cohort. This figure is fivefold
vious exposure to anti-TB treatment as a common risk higher than the expected estimate by the WHO, which
Kebede and Mamo BMC Pulmonary Medicine (2024) 24:527 Page 8 of 10

should be less than 10% [48]. Additionally, the current Conclusion

finding is much higher than the national estimates of The findings highlight the challenges and gaps in the
MDR-TB patients with unfavorable treatment out- management of MDR-TB in Ethiopia. The high propor-
comes [49, 50]. The observed discrepancy might be tion of previously treated cases is likely a re-infection
due to variations in sample sizes, socio-demographic after cure from the previous, and the emergence of
factors, and clinical conditions of the participants. additional resistance to SLDs are concerning challenge
The relatively higher proportions of patients with poor to the effectiveness of the program control. The treat-
treatment outcomes may increase the risk of additional ment success rate falls short of the WHO target (≥ 90).
developments of MDR-TB. Addressing this gap in TB The death rate among patients, particularly those with
control is costly and has an impact on the quality of additional resistance to SLD, and LTFU is alarming and
life, as it may lead to individuals developing XDR-TB. calls for improved patient-centered services. The pres-
Thus, enhanced monitoring and targeted evaluation of ence of MDR-TB cases with no assigned treatment out-
patients are necessary to save lives. comes reflects the limitations in laboratory access and
Additionally, in the present study, a significant num- the need for better diagnostic tools. Addressing these
ber of MDR-TB cases (28.9%) were found to have no issues requires increased resources, improved labora-
determined records of treatment outcome after having tory capacity, and the implementation of integrated and
considerable treatment period in the meantime. The comprehensive care for MDR-TB patients, particularly
reason for not evaluating and determining the patients’ those with co-morbidities such as HIV and those who are
status early might be due to a lack of access to reliable underweight. Ultimately, a multi-sectored approach and
laboratory testing. For example, 89.9% of cases in the strong commitment from the government and stakehold-
current study were confirmed using GeneXpert, which ers are crucial to effectively manage, control and elimi-
only detects MTB and RIF’s resistances [51]. There- nate MDR-TB in Ethiopia.
fore, improved laboratory access should be considered
in the future to the greatest extent possible in the area.
In Ethiopia, like any other sub-Saharan African coun- Limitations
try, MDR-TB detection rates are under-reported due to This study has some limitations. It relied on exist-
factors such as a lack of reliable, quick, and affordable ing medical records, which may have had incomplete
diagnostic testing [52]. This not only affects treatment or missing data on important variables such as patient
outcomes but also hinders effective infection control demographics, clinical characteristics, treatment regi-
and the implementation of evidence-based measures for mens, and outcomes. The patients included in the retro-
MDR-TB control. spective cohort may not have been representative of the
Our study identified HIV co-infection, and low BMI overall MDR-TB population, as they were selected based
as significant predictors for unfavorable treatment out- on availability of data, which can introduce selection bias.
comes in MDR-TB patients. Several previous reports The study may not have been able to adequately control
from Ethiopia [45, 47, 53–56] have shown HIV co- for all potential confounding factors, such as comorbidi-
infection as a significant predictor of poor treatment ties, disease severity, access to healthcare, socioeconomic
outcomes, highlighting the need for integrated TB-HIV and genetic factors, which can influence treatment out-
services and comprehensive care for patients with both comes. Additionally, the data may not reflect the current
conditions. A similar retrospective study from Ethiopia treatment landscape, as the management of MDR-TB has
[45] has shown a significant association between low evolved over time. The study also had limited informa-
BMI and unfavorable MDR-TB treatment outcome. The tion on long-term outcomes and was unable to account
presence of non-HIV comorbidities also suggests a trend for uncontrolled factors, such as patient adherence to
towards unfavorable outcomes, though further research treatment or the impact of social determinants of health.
may be needed to clarify this relationship. Addressing These limitations can affect the internal and external
nutritional status and managing HIV co-infection could validity of the study findings, and the results should be
be critical in improving treatment outcomes in this pop- interpreted with caution when making decisions about
ulation. On the other hand, high impact studies like [47] clinical practice or policy.
showed that direct contact with known patients with TB, Abbreviations
history of cigarette smoking and living in rural were sig- AIDS Acquired Immunodeficiency Syndrome
AOR Adjusted Odds Ratio
nificant risk factors for lower treatment outcome among
BMI Body-Mass-Index
MDR-TB patients. CI Confidence Interval
COR Crude Odds Ratio
CNS-IRB College of Natural and Computational Sciences Institutional
Review Board
Kebede and Mamo BMC Pulmonary Medicine (2024) 24:527 Page 9 of 10

DOT Directly Observed Treatment References

DR-TB Drug Resistant Tuberculosis 1. WHO. Global tuberculosis report 2023. Geneva: World Health Organiza-
DST Drug Susceptibility Test tion; 2023. Licence: CC BY NC-SA 3.0 IGO.
FLDs First-Line Drugs 2. https://​www.​cdc.​gov/​tb/​about/​drug-​resis​tant.​html. Accessed 3 Aug
FDRE Federal Democratic Republic of Ethiopia 2024.
FMoH Federal Ministry of Health of Ethiopia 3. Mnyambwa NP, Philbert D, Kimaro G, et al. Gaps related to screening and
FQs Fluoroquinolone diagnosis of tuberculosis in care cascade in selected health facilities in
HIV Human Immunodeficiency Virus East Africa countries: a retrospective study. J Clin Tuberc Other Mycobact
INH Isoniazid Dis. 2021;25:100278.
IQR Inter Quartile Range 4. Muthukrishnan L. Multidrug resistant tuberculosis - Diagnostic challenges
LPA Line-Probe Assay and its conquering by nanotechnology approach - An overview. Chem
MDR-TB Multidrug-Resistant Tuberculosis Biol Interact. 2021;337: 109397.
MTB  Mycobacterium tuberculosis 5. Singh S, Dey B, Sachdeva KS, Kabra SK, Chopra KK, Chaudhary VK, Sharma
NA Not applicable P, Katoch VM. Challenges in tuberculosis diagnosis and management:
NTLCP National Tuberculosis and Leprosy Control Program recommendations of the expert panel. J Lab Physicians. 2015;7(1):1–3.
RR/MDR-TB Rifampicin Resistant or Multidrug-resistant Tuberculosis 6. Danchuk SN, Solomon OE, Kohl TA, et al. Challenging the gold standard:
SLDs Second Line Drugs the limitations of molecular assays for detection of Mycobacterium
SNRS Sidama National Regional State tuberculosis heteroresistance. Thorax. 2024;79(7):670–5. https://​doi.​org/​
SEPR Southern Ethiopian Peoples’ Region 10.​1136/​thorax-​2023-​220202.
SPSS Statistical Package for Social Sciences 7. Getahun M, Ameni G, Mollalign H, Diriba G, Beyene D. Genotypic and
TB Tuberculosis phenotypic drug-resistance detection and prevalence of heteroresist-
TB/HIV Tuberculosis and/or Human Immunodeficiency Virus ance in patients with isoniazid- and multidrug-resistant tuberculosis in
TIC Treatment Initiating Centre Ethiopia. IJID Regions. 2022;2:149–53.
YGH Yirgalem General Hospital 8. Mohammed H, Oljira L, Roba KT, et al. Burden of tuberculosis and chal-
WHO World Health Organization lenges related to screening and diagnosis in Ethiopia. J Clin Tuberc Other
XDR-TB Extensively Drug Resistant Tuberculosis Mycobact Dis. 2020;19:100158.
9. WHO. World Health Organization consolidated guidelines on tubercu-
Acknowledgements losis: module 4: treatment: drug-resistant tuberculosis treatment, 2022
The authors would like to acknowledge the support of Yirgalem Medical Col- update. Geneva: World Health Organization. Available at https://​www.​
lege staff for their kind support during data collection. who.​int/​publi​catio​ns/i/​item/​97892​40063​129, Accessed 25 Aug 2024.
10. Jiang W. Addressing the adherence challenge in tuberculosis treatment:
Authors’ contributions more than digital technologies. Lancet Glob Health. 2023;11(5):E634–635.
AHK and HM conceived the research idea, AHK collected the data and 11. Cáceres G, Calderon R, Ugarte-Gil C. Tuberculosis and comorbidities:
drafted the manuscript, HM critically revised the the manuscript. Both authors treatment challenges in patients with comorbid diabetes mellitus and
reviewed and approved the final version of the manuscript. depression. Ther Adv Infect Dis. 2022;9:20499361221095830.
12. WHO consolidated guidelines on tuberculosis. Module 6: tuberculosis
Funding and comorbidities. Geneva: World Health Organization; 2024. Licence: CC
This study did not have a specific funding source. BY-NC-SA 3.0 IGO.
13. Duarte R, Lönnroth K, Carvalho C, et al. Tuberculosis, social determinants
Data availability and co-morbidities (including HIV). Pulmonol. 2018;24(2):115–9.
The datasets used and/or analysed during this study are available from the 14. Wagnew F, Alene KA, Kelly M, Gray D. Undernutrition increases the risk of
corresponding author upon reasonable request. unsuccessful treatment outcomes of patients with tuberculosis in Ethio-
pia: A multicenter retrospective cohort study. J Infect. 2024;89(1): 106175.
15. Khurana A, Dhingra B. The diagnosis and challenges of pediatric tuber-
Declarations culosis. In: Singh A A, Sharma D, editors. Diagnosis of Mycobacterium.
Singapore: Springer; 2023.
Ethics approval and consent of participants 16. Basile FW, Nabeta P, Ruhwald M, Song R. Pediatric tuberculosis diagnos-
The study received approval from the College of Natural and Computational tics: Present and future. JPIDS. 2022;11(Suppl 3):S85–93.
Sciences Institutional Review Board (CNS-IRB), Addis Ababa University (Ref 17. WHO. Roadmap for childhood tuberculosis: towards zero deaths. Geneva:
No: CNSDO/605/13/2021). Additionally, official permission was obtained from World Health Organization; 2013. WHO/HTM/TB/2013.12.
YGH prior to commencing the study. As the study had a retrospective design, 18. WHO. Paediatric drug optimization for tuberculosis: meeting report,
participant consent was not applicable and the CNS-IRB waived the need for October 2023. Geneva: World Health Organization; 2024. Licence: CC BY-
consent. However, to ensure the confidentiality of the patients’ medical his- NC-SA 3.0 IGO.
tory, each record was assigned a unique identifier code. 19. Lönnroth K, Jaramillo E, Williams BG, Dye C, Raviglione M. Drivers of tuber-
culosis epidemics: The role of risk factors and social determinants. Soc Sci
Consent for publication Med. 2009;68:2240–6.
Not applicable. 20. Lönnroth K, Castro K, Chakaya JM, et al. Tuberculosis control and
elimination 2010–50: cure, care, and social development. Lancet.
Competing interests 2010;375:1814–29.
The authors declare no competing interests. 21. Nsubuga P, White ME, Thacker SB, et al. Public Health Surveillance: A
tool for targeting and monitoring interventions. In: Jamison DT, Breman
Author details JG, Measham AR, et al., editors. Disease Control Priorities in Develop-
1
Department of Medical Microbiology, Yirgalem General Hospital Medical ing Countries. 2nd edition. Washington (DC): The International Bank for
College, PO Box 184, Yirgalem, Ethiopia. 2 Department of Microbial, Cellular Reconstruction and Development/The World Bank; 2006. Chapter 53.
and Molecular Biology, College of Natural and Computational Sciences, Addis Available from: https://​www.​ncbi.​nlm.​nih.​gov/​books/​NBK11​770/ Co-
Ababa University, PO Box 1176, Addis Ababa, Ethiopia. published by Oxford University Press, New York.
22. WHO. Guidelines for surveillance of drug resistance in tuberculosis,
Received: 10 February 2024 Accepted: 17 October 2024 5th edition. World Health Organization Document 2015; WHO/HTM/
Kebede and Mamo BMC Pulmonary Medicine (2024) 24:527 Page 10 of 10

TB/2015.13: 1–63. Available at https://​www.​who.​int/​publi​catio​ns-​detail-​ among adult smear positive patients in Eastern Ethiopia. Infect Drug
redir​ect/​97892​41549​134. Accessed 4 Aug 2024. Resist. 2021;14:4493–500.
23. Schwoebel V. Surveillance of tuberculosis. Indian J Tuberc. 45. Kassa GM, Tadesse A, Gelaw YA, et al. Predictors of mortality among
2020;67(4S):S33–42. multidrug-resistant tuberculosis patients in central Ethiopia: a retrospec-
24. Dean AS, Tosas Auguet O, Glaziou P, et al. 25 years of surveillance of drug- tive follow-up study. Epidemiol Infect. 2020;148: e258.
resistant tuberculosis: achievements, challenges, and way forward. Lancet 46. Bogale L, Tsegaye T, Abdulkadir M, Akalu TY. Unfavorable treatment
Infect Dis. 2022;22(7):e191–6. outcome and its predictors among patients with multidrug-resistance
25. https://​www.​who.​int/​teams/​global-​tuber​culos​is-​progr​amme/​tb-​repor​ts/​ tuberculosis in southern Ethiopia in 2014 to 2019: A multi-center retro-
global-​tuber​culos​is-​report-​2023/​tb-​disea​se-​burden/​1-4-​tb-​preva​lence-​ spective follow-up study. Infect Drug Resist. 2021;14:1343–55.
surve​ys, 4 Aug 2024. 47. Admassu F, Abera E, Gizachew A, et al. Risk factors of multidrug resistant
26. ESS. Population six by Sex, Zone and Woreda July 2023. Addis Ababa: tuberculosis among patients with tuberculosis at selected multidrug
Ethiopia Statistical Service; 2023. resistance treatment initiative centres in southern Ethiopia: a case-control
27. Gebrekidan G, Tesfaye G, Hambisa MT, et al. Quality of tuberculosis care study. BMJ Open. 2023;13: e061836.
in private health facilities of Addis Ababa. Ethiopia Tuberc Res Treat. 48. WHO. Global Tuberculosis report. Geneva: World Health Organization;
2014;2014: 720432. 2021.
28. Mosissa L, Kebede A, Mindaye T, et al. External quality assessment of 49. Molie T, Teklemariam Z, Klinkenberg E, et al. Intensive phase treatment
AFB smear microscopy performances and its associated factors in outcome and associated factors among patients treated for multidrug
selected private health facilities in Addis Ababa. Ethiopia Pan Afr Med J. resistant tuberculosis in Ethiopia: a retrospective cohort study. BMC Infect
2016;24:125. Dis. 2019;19(1):818.
29. WHO consolidated guidelines on tuberculosis. Module 3: Diagnosis – 50. Tola H, Holakouie- Naieni K, Mansournia MA, et al. National treatment
rapid diagnostics for tuberculosis detection 2021 update. Geneva: World outcome and predictors of death and treatment failure in multidrug-
Health Organization; 2021 . https://​www.​who.​int/​publi​catio​ns/i/​item/​ resistant tuberculosis in Ethiopia: a 10-year retrospective cohort study.
97892​40029​415. BMJ Open. 2021;11: e040862.
30. Definitions and reporting framework for tuberculosis – 2013 revision 51. http://​www.​who.​int/​media​centre/​news/​relea​ses/​2010/​tb_​test_​20101​
(updated December 2014) (WHO/HTM/TB/2013.2). Geneva: World Health 208/​en/​index.​html, retrieved 14 Aug 2024.
Organization; 2013. https://​apps.​who.​int/​iris/​bitst​ream/​handle/​10665/​ 52. Chin DP, Hanson CL. Finding the missing tuberculosis patients. J Infect
79199/​97892​41505​345_​eng.​pdf. Dis. 2017;216(suppl_7):S675–8.
31. FMoH. National Guidelines for Tuberculosis, Leprosy and TB/HIV Co- 53. Girum T, Muktar E, Lentiro K, Wondiye H, Shewangizaw M. Epidemiology
infection Prevention and Control. Addis Ababa: Rederal Ministry of health; of multidrug-resistant tuberculosis (MDR-TB) in Ethiopia: a systematic
2017. review and meta-analysis of the prevalence, determinants and treatment
32. WHO. Algorithm for laboratory diagnosis and treatment-monitoring of outcome. Trop Dis Travel Med Vaccines. 2018;4:5.
pulmonary tuberculosis and drug-resistant tuberculosis using state-of- 54. Ketema DB, Muchie KF, Andargie AA. Time to poor treatment outcome
the-art rapid molecular diagnostic technologies Expert opinion of the and its predictors among drug-resistant tuberculosis patients on second-
European Tuberculosis Laboratory Initiative core group members for the line antituberculosis treatment in Amhara region, Ethiopia: retrospective
WHO European Region. 2017. cohort study. BMC Public Health. 2019;19:1481.
33. Gotham D, McKenna L, Deborggraeve S, et al. Public investments in the 55. Alemu AA, Bitew ZW, Worku T. Poor treatment outcome and its predictors
development of GeneXpert molecular diagnostic technology. PLoS ONE. among drug-resistant tuberculosis patients in Ethiopia: A systematic
2021;16(8): e0256883. review and meta-analysis. Int J Infect Dis. 2020;98:420–39.
34. Padda IS, Muralidhara Reddy K. Antitubercular medications. [Updated 56. Tola H, Holakouie-Naieni K, Mansournia MA, et al. National treatment
2023 Jun 3]. In: StatPearls. Treasure Island (FL): StatPearls Publishing; 2024. outcome and predictors of death and treatment failure in multidrug-
Available from: https://​www.​ncbi.​nlm.​nih.​gov/​books/​NBK55​7666/. resistant tuberculosis in Ethiopia: a 10-year retrospective cohort study.
35. Jang JG, Chung JH. Diagnosis and treatment of multidrug-resistant tuber- BMJ Open. 2021;11: e040862.
culosis. Yeungnam Univ J Med. 2020;37(4):277–85.
36. Wáng YXJ, Chung MJ, Skrahin A, et al. Radiological signs associated with
pulmonary multi-drug resistant tuberculosis: an analysis of published Publisher’s Note
evidences. Quant Imaging Med Surg. 2018;8(2):161–73. Springer Nature remains neutral with regard to jurisdictional claims in pub-
37. Opollo VS, Wu X, Hughes MD, et al. HIV testing uptake among the house- lished maps and institutional affiliations.
hold contacts of multidrug-resistant tuberculosis index cases in eight
countries. Int J Tuberc Lung Dis. 2018;22(12):1443–9.
38. Definitions WHO, reporting framework for tuberculosis -,. revision
(Updated December 2014 and January 2020). Geneva: World Health
Organization; 2013.
39. FMOH. Guidelines for Clinical and Programmatic Management of TB, TB/
HIV, DR-TB and Leprosy in Ethiopia. 7th ed. Addis Ababa: MOH; 2021. p.
79–93.
40. Eshetie S, Alebel A, Wagnew F, Geremew D, Fasil A, Sack U. Current treat-
ment of multidrug resistant tuberculosis in Ethiopia: an aggregated and
individual patients’ data analysis for outcome and effectiveness of the
current regimens. BMC Infect Dis. 2018;18(1):486.
41. Belachew T, Yaheya S, Tilahun N, et al. Multidrug-resistant tuberculosis
treatment outcome and associated factors at the University of Gondar
Comprehensive Specialized Hospital: A ten-year retrospective study.
Infect Drug Resist. 2022;15:2891–9.
42. Biadglegne F, Sack U, Rodloff AC. Multidrug-resistant tuberclosis in Ethio-
pia: efforts to expand diagnostic services, treatment and care. Antimicrob
Resist Infect Control. 2014;3(1):31.
43. Eshetie S, Gizachew M, Dagnew M, et al. Multidrug resistant tuberculosis
in Ethiopian settings and its association with previous history of anti-
tuberculosis treatment. BMC Infect Dis. 2017;17(1):219.
44. Amin Z, Mitiku H, Marami D, Shume T, Weldegebreal F. Magnitude of
multidrug resistance and associated factors of pulmonary tuberculosis