Kraef et al. BMC Infect Dis (2021) 21:1038 https://doi.org/10.1186/s12879-021-06745-w RESEARCH Open Access Delayed diagnosis of tuberculosis in persons living with HIV in Eastern Europe: associated factors and effect on mortality—a multicentre prospective cohort study Christian Kraef1,2,14*, Adrian Bentzon1, Alexander Panteleev3, Alena Skrahina4, Natalie Bolokadze5, Simona Tetradov6, Regina Podlasin7, Igor Karpov8, Elena Borodulina9, Elena Denisova10, Inga Azina11, Jens Lundgren1, Isik Somuncu Johansen12, Amanda Mocroft1,13, Daria Podlekareva1 and Ole Kirk1,2,12for the T. B.:H. I. V. Study Group Abstract Background: Early diagnosis of tuberculosis (TB) is important to reduce transmission, morbidity and mortality in people living with HIV (PLWH). Methods: PLWH with a diagnosis of TB were enrolled from HIV and TB clinics in Eastern Europe and followed until 24 months. Delayed diagnosis was defined as duration of TB symptoms (cough, weight-loss or fever) for ≥ 1 month before TB diagnosis. Risk factors for delayed TB diagnosis were assessed using multivariable logistic regression. The effect of delayed diagnosis on mortality was assessed using Kaplan–Meier estimates and Cox models. Findings: 480/740 patients (64.9%; 95% CI 61.3–68.3%) experienced a delayed diagnosis. Age ≥ 50 years (vs. < 50 years, aOR = 2.51; 1.18–5.32; p = 0.016), injecting drug use (IDU) (vs. non-IDU aOR = 1.66; 1.21–2.29; p = 0.002), being ART naïve (aOR = 1.77; 1.24–2.54; p = 0.002), disseminated TB (vs. pulmonary TB, aOR = 1.56, 1.10–2.19, p = 0.012), and presenting with weight loss (vs. no weight loss, aOR = 1.63; 1.18–2.24; p = 0.003) were associated with delayed diagnosis. PLWH with a delayed diagnosis were at 36% increased risk of death (hazard ratio = 1.36; 1.04–1.77; p = 0.023, adjusted hazard ratio 1.27; 0.95–1.70; p = 0.103). Conclusion: Nearly two thirds of PLWH with TB in Eastern Europe had a delayed TB diagnosis, in particular those of older age, people who inject drugs, ART naïve, with disseminated disease, and presenting with weight loss. Patients with delayed TB diagnosis were subsequently at higher risk of death in unadjusted analysis. There is a need for optimi- sation of the current TB diagnostic cascade and HIV care in PLWH in Eastern Europe. Introduction Globally, about 10 million people were diagnosed with tuberculosis (TB) in 2019 [1]. Among those, 8.2% were people living with the human immunodeficiency virus (PLWH) [1]. Among PLWH, TB is the most frequent co- *Correspondence: christiankraef@gmail.com infection, cause for hospitalization and cause of death, 1 CHIP (Centre of Excellence for Health, Immunity and Infections), responsible for about 208,000 AIDS-related deaths Rigshospitalet, University of Copenhagen, Copenhagen, Denmark Full list of author information is available at the end of the article worldwide in 2019 [1, 2]. One major challenge to improve © The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, 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 changes were made. 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://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. The Creative Commons Public Domain Dedication waiver (http://​creat​iveco​ mmons.​org/​publi​cdoma​in/​zero/1.​0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Kraef et al. BMC Infect Dis (2021) 21:1038 Page 2 of 12 survival for TB patients is to ensure timely diagnosis of and TB clinics in nine Eastern European countries TB [3]. In patients mono-infected with TB a recent sys- (Belarus, Estonia, Georgia, Latvia, Lithuania, Poland, tematic review of delay in diagnosis of pulmonary TB Romania, Russia and Ukraine) between Jan 1, 2011 in low- and middle income countries found that 42% of and Dec 31, 2013 [17]. All methods were carried out in TB patients had a delay between symptom onset and TB accordance with relevant guidelines and regulations. diagnosis of at least a month [4]. A systematic review of All participating clinics obtained ethical approval in post-mortem studies from Africa, Asia and the Ameri- accordance with local rules and legislations. A list of cas found TB as the cause of death in 37.2% of PLWH, IEC/IRB from each participating center is provided in of which 45.8% remained undiagnosed at the point of Appendix. Informed consent was obtained from all par- death [3]. Delayed diagnosis of TB has repeatedly been ticipants. The study was performed in accordance with shown to lead to more severe disease presentation, and the STROBE guidelines for observational studies [18]. increases morbidity and mortality significantly [5–11]. Furthermore, delayed TB diagnosis increases the risk for TB community transmission [6, 12]. Variables, inclusion and exclusion criteria In consequence, one component of the World Health Explanatory variables include demography data, previ- Organization’s End TB strategy to reduce TB deaths by ous TB-disease, type of TB diagnosis, clinical presenta- 75% and TB incidence by 50% in the general population tion of current TB, HIV characteristics including CD4 by the year 2025 is ensuring early diagnosis of TB [13]. cell counts and HIV-RNA measurements, antiretroviral To this end, the strategy identifies amongst others the therapy (ART) history and AIDS defining diagnoses. mapping of high-risk groups and the implementation of Variables with missing information on treatment with carefully planned systematic screening for active disease cotrimoxazole, chest x-ray, HBsAg, Anti-HCV and TB among identified high-risk groups as important facili- risk factors (alcohol, previous TB, imprisonment) were tators of early diagnosis [13]. One high-risk group are assumed as absent. Those with missing information on PLWH, in whom the effect of delayed diagnosis on mor- CD4 count were included as a separate category. For bidity and mortality is even more distinct compared to drug-susceptibility testing and determination of MDR- those mono-infected with TB [14–16]. TB, if some information on drug-susceptibility was pre- HIV/TB-coinfections continue to be a major chal- sent, but missing for a specific drug M. tuberculosis was lenge in many Eastern European countries [17]. In a large assumed to be susceptible to that specific drug, as done clinical cohort of patients with HIV/TB across Eastern previously [19]. Europe, 1-year mortality was 27%, with the majority of To evaluate the delayed diagnosis, we used the varia- those deaths (79%) directly attributable to TB [17]. That ble “symptom duration” prior to diagnosis that was cap- is a risk of death nearly four-times higher than that in tured as a categorical variable (< 1 month, 1–3 months PLWH with TB from western Europe and Latin America and > 3 months) in the original enrollment form by the [17]. enrolling health care worker based on self-reporting by However, to the best of our knowledge, the contribu- the PLWH. For the purpose of this study, we merged tion of delayed TB diagnosis to morbidity and mortality the categories to create a binary variable (≥ 1 months and its associated factors in PLWH have not been inves- versus < 1 months). tigated in an Eastern European setting. The objective of this study is to describe the prevalence of delayed diagno- Box 1. Study definitions sis, analyze factors associated with delayed diagnosis and Region Eastern Europe Belarus, Estonia, Georgia, to quantify the effect of delayed diagnosis on survival in Latvia, Lithuania, Poland, PLWH. Romania, Ukraine, Russia Delayed diagnosis Delayed diagnosis was defined as duration of TB Methods symptoms (cough, weight- Study design, population and sample loss or fever) for ≥ 1 month before TB diagnosis The data for this analysis were drawn form an interna- tional prospective observational cohort study (TB:HIV TB diagnosis Definite Positive culture or PCR (Xpert) for Mtb (in sputum study). Details of the study have previously been pub- smear or other specimen) lished [17]. Study protocols and forms are available at www.​chip.​dk. Only patients included at Eastern Euro- pean study sites were included in the current analysis. The TB:HIV study enrolled consecutive HIV-patients with TB diagnosis aged 16 years or older from 21 HIV Kraef et al. BMC Infect Dis (2021) 21:1038 Page 3 of 12 Probable Acid fast bacilli or time (0–2 months and 2–24 months). The proportional granulomatous inflamma- hazard assumption was tested based on Schoenfeld resid- tion in sputum smear or tissue biopsy and other uals. Stata IC 15.1 was used for statistical analysis. specimens (e.g., CSF, Pleural fluid) Results Presumptive A patient who presents In the TB:HIV cohort 825 PLWH from Eastern Europe with symptoms or signs suggestive of TB where TB were eligible for the analysis, of those 85 (10.3%) were treatment is initiated and excluded because of missing data on symptom duration. not subsequently stopped A total of 740 patients from Eastern Europe were because the TB diagnosis was ruled out included in the current analysis. The analysis included TB location Pulmonary TB localised to the lungs, 260 (35.1%) patients with a delay of less than 1 month larynx, or tracheobronchial and 480 (64.9%) patients with a delay of at least 1 month, tree 380 (51.4%) had a delay of 1–3 months, and 100 (13.5%) Extrapulmonary just one extra-pulmonary of > 3 months. site was identified Disseminated Either of the following: (i) TB documented in at Factors associated with delayed diagnosis least two organ systems Baseline characteristics of the TB:HIV patients in East- (one of which could be ern Europe stratified by delayed diagnosis, and fac- lungs) (ii) miliary TB, or tors associated with delayed diagnosis are presented in (iii) isolation of Mtb from Table 1. In the multivariable analysis age ≥ 50 years (vs. blood or bone marrow < 50 years, aOR 2.51, 95% CI 1.18–5.32, p = 0.016), Inject- TB drug resistance MDR Mtb resistant to both ing Drug Use (IDU) (vs. non-IDU aOR 1.66, 95% CI rifamycin and isoniazid 1.21–2.29, p = 0.002), being ART treatment naïve at TB AIDS Clinical AIDS criteria (ref: 1993 Revised Classification diagnosis (vs. any ART before TB diagnosis, aOR 1.77, System for HIV Infection 95% CI 1.24–2.54, p = 0.002), disseminated TB (vs. pul- and Expanded Surveillance monary TB, aOR 1.56, 95% CI 1.10–2.19, p = 0.012), and Case Definition for AIDS Among Adolescents and weight loss (vs. no weight loss, aOR 1.63, 95% CI 1.18– Adults) 2.24, p = 0.003) were associated with delayed diagnosis (≥ 1 month). Conversely, a previous TB diagnosis (vs. never diagnosed with TB, aOR 0.60, 95% CI 0.38–0.95, Data analysis p = 0.029) was associated with earlier diagnosis. Descriptive statistics were used to present data as pro- In an analysis of the organ systems affected, i.e. organs portions, medians and interquartile ranges. The Chi- where TB was identified during diagnostic work-up squared and Fisher’s exact test were used to compare (non-exclusive/several sites possible in the same patient) categorical variables between groups, while the Kruskal– genitourinary (OR 3.64; 95% CI 1.39–9.52; p = 0.006) and Wallis test was used to compare continuous variables. lymphatic TB (OR 2.28; 95% CI 1.65–3.17; p = 0.001) To identify risk factors for delayed diagnosis we first were associated with delayed diagnosis, whereas pulmo- conducted a bivariate analysis of potential risk factors. nary TB was not (OR 1.17; 95% CI 0.74–1.86, p = 0.504). Those with a p ≤ 0.10 were included in the multivariable logistic regression model. Survival analysis For the survival analysis baseline was defined as the Survival over 24 months, stratified by delayed diagno- date of TB treatment initiation. All follow-up was cen- sis, is presented in the Kaplan–Maier analysis in Fig. 1. sored at 24 months. Overall survival comparing the group Estimates of all-cause mortality at 24 months were 262 with delayed and non-delayed diagnosis was assessed (35.4%), 80 (30.8%) in the early diagnosis group and 182 using the Kaplan–Meier (KM) method and the log-rank (37.9%) in the late diagnosis group. In the first 2 months test. We used Cox proportional hazards regression mod- of follow-up, the all-cause mortality rate did not dif- els to estimate hazard ratios (HRs) and 95% confidence fer between PLWH with and without delayed diagno- intervals (CIs) of the effect of delayed diagnosis on mor- sis, (100/100-person-years of follow-up (PYFU) (95% tality. Variables with a p-value ≤ 0.10 were included in the CI 78–127) v. 91/100 PYFY, respectively, log-rank test multivariate model to assess the hazard ratio (HR) and p = 0.566. For the follow-up time from 2 to 24 months all- the Cox-model was stratified by the clinical center. Based cause mortality rates were 23/100 PYFU (95% CI 19–28) on the Kaplan–Maier survival analysis, we decided to and 14/100 PYFU (95% CI 11–19), respectively, log-rank stratify the Cox-regression model at 2 months follow-up test p = 0.011. Kraef et al. BMC Infect Dis (2021) 21:1038 Page 4 of 12 Table 1 Baseline characteristics of the TB:HIV patients in Eastern Europe stratified by delayed diagnosis and factors associated with delayed diagnosis Total Delayed diagnosis OR (95% CI) p-value Adjusted OR (95% CI) p-value < 1 month ≥ 1 month n = 740 n = 260 n = 480 Gender Male 563 (76.1) 200 (76.9) 363 (75.6) 0.93 (0.65–1.33) 0.69 Age Age 16–49 694 (93.8) 250 (96.2) 444 (92.5) Ref Ref Ref Ref Age ≥ 50 46 (6.2) 10 (3.8) 36 (7.5) 2.62 (1.21–5.66) 0.01 2.51 (1.18–5.32) 0.016 Exposure group (HIV) MSM (yes vs no) 10 (1.4) 5 (1.9) 5 (1.0) 0.54 (0.15–1.87) 0.32 IDU (yes vs no) 422 (57.0) 132 (50.8) 290 (60.4) 1.48 (1.09–2.00) 0.01 1.66 (1.21–2.29) 0.002 Heterosexual (yes vs no) 183 (24.7) 72 (27.7) 111 (23.1) 0.79 (0.56–1.11) 0.17 Known HIV at TB diagnosis 668 (90.3) 235 (90.4) 433 (90.2) 0.98 (0.59–1.63) 0.94 Treatment history (HIV) ART naïve at TB ­diagnosisa 558 (75.4) 176 (67.7) 382 (79.6) 1.86 (1.32–2.63) 0.0001 1.77 (1.24–2.54) 0.002 ART use at TB ­diagnosisa 132 (17.8) 59 (22.7) 73 (15.2) 0.61 (0.42 –0.9) 0.011 Omittedd Cotrimoxazole at TB diagnosis 273 (36.9) 84 (32.3) 189 (39.4) 1.36 (0.99–1.87) 0.057 1.25 (0.89–1.74) 0.197 CD4-cell count/mm3° 0–199/mm3 440 (59.5) 151 (58.1) 289 (70.8) 1.06 (0.74–1.52) 0.745 ≥ 200 185 (25.0) 66 (25.4) 119 (29.2) Ref Ref Missing CD4 115 (15.5) 43 (37.4) 72 (62.6) 0.93 (0.57–1.51) 0.764 Prior AIDS 177 (23.9) 66 (25.4) 111 (23.1) 0.88 (0.62–1.26) 0.49 Previous TB Yes vs no 99 (13.4) 46 (17.7) 53 (11.0) 0.58 (0.38–0.89) 0.011 0.60 (0.38–0.95) 0.029 TB risk factor Alcohol abuse (yes vs no) 186 (25.1) 72 (27.7) 114 (23.8) 0.81 (0.58–1.15) 0.24 Recent TB in family (yes vs no) 53 (7.2) 17 (6.5) 36 (7.5) 1.16 (0.64–2.11) 0.63 Imprisonment within 2 years (yes vs no) 143 (19.3) 47 (18.1) 96 (20.0) 1.13 (0.77–1.67) 0.53 Clinical presentation of TB Pulmonary (including trachea and larynx) 239 (32.3) 101 (38.9) 138 (28.8) Ref Ref Ref Ref Extrapulmonary 54 (7.3) 23 (8.9) 31 (6.5) 0.99 (0.54–1.79) 0.964 0.83 (0.44–1.55) 0.553 Disseminated 447 (60.4) 136 (52.3) 311 (64.8) 1.67 (1.21–2.32) 0.002 1.56 (1.10–2.19) 0.012 Localisations TB was found (non-exclusive) Pulmonary (including trachea and larynx) 654 (88.4) 227 (87.3) 427 (89.0) 1.17 (0.74–1.86) 0.504 Omittede Pleural 92 (12.4) 32 (12.3) 60 (12.5) 1.02 (0.64–1.61) 0.94 Omittede Lymphatic 309 (41.8) 76 (29.2) 233 (48.5) 2.28 (1.65–3.17) 0.0001 Omittede Bone and/or joint 13 (1.8) 2 (0.8) 11 (2.3) 3.03 (0.66–13.8) 0.155 Omittede Genitourinary 37 (5.0) 5 (1.9) 32 (6.7) 3.64 (1.39–9.52) 0.004 Omittede Meningeal and other central nervous system 74 (10.0) 28 (10.8) 46 (9.6) 0.88 (0.53–1.44) 0.608 Omittede Gastro-intestinal 55 (7.4) 16 (6.2) 39 (8.1) 1.35 (0.74–2.46) 0.329 Omittede TB symptoms Cough (yes vs no) 501 (67.7) 188 (72.3) 313 (65.2) 0.72 (0.52–0.99) 0.049 0.72 (0.51–1.02) 0.061 Fever (yes vs no) 636 (86.0) 227 (87.3) 409 (85.2) 0.84 (0.54–1.31) 0.433 Weight loss (yes vs no) 445 (60.1) 133 (51.2) 312 (65.0) 1.77 (1.3–2.42) 0.0001 1.63 (1.18–2.24) 0.003 Number of ­symptomsc None 20 (2.7) 6 (2.3) 14 (2.9) Ref Ref One symptom 133 (18.0) 50 (19.2) 83 (17.3) 0.71 (0.26–1.97) 0.512 Two symptoms 312 (42.2) 114 (43.9) 198 (41.3) 0.74 (0.28–2.0) 0.556 Three symptoms 275 (37.2) 90 (34.6) 185 (38.5) 0.88 (0.33–2.37) 0.802 Kraef et al. BMC Infect Dis (2021) 21:1038 Page 5 of 12 Table 1 (continued) Total Delayed diagnosis OR (95% CI) p-value Adjusted OR (95% CI) p-value < 1 month ≥ 1 month n = 740 n = 260 n = 480 Screening chest X-ray before current episode Yes vs no 338 (45.7) 115 (44.2) 223 (46.5) 1.09 (0.81–1.48) 0.56 Rifampicin resistance Yes vs no 95 (38.3) 34 (35.8) 61 (39.9) 1.19 (0.7–2.0) 0.52 MDR-TB Yes 88 (11.9) 31 (11.9) 57 (11.9) 1.22 (0.71–2.11) 0.478 No 148 (20.0) 59 (22.7) 89 (18.5) Ref Ref Missing drug susceptibility testing 504 (68.1) 170 (65.4) 334 (69.6) 1.30 (1.09–1.90) 0.170 Diagnosis Definite 333 (45.0) 123 (47.3) 210 (43.8) Ref Ref Probable 76 (10.3) 24 (9.2) 52 (10.8) 1.27 (0.75–2.16) 0.380 Presumptive 331 (44.7) 113 (43.5) 218 (45.4) 1.13 (0.82–1.55) 0.451 Hepatitis B ­ aselineb HbsAg at b 44 (6.0) 17 (6.5) 27 (5.6) 0.85 (0.46–1.59) 0.616 Hepatitis C Anti-HCV ­antibodiesb 405 (54.7) 134 (51.5) 271 (56.5) 1.21 (0.9–1.65) 0.199 Continuous variables Weight (kg)   Baseline 60 (53–68) 63 (55–68.5) 60 (53–68) 0.99 (0.97–1.01) 0.255 HIV-RNA (log10) copies/ml   Baseline 5.3 (4.5–5.8) 5.2 (4.3–5.7) 5.3 (4.6–5.8) 1.12 (0.96–1.32) 0.158 Adjusted for age, IDU, ART treatment naivety, cotrimoxazole treatment, previous TB, clinical TB presentation and TB symptoms cough and weight loss OR odds ratio, CI confidence interval, MSM Men Who Have Sex with Men, IDU injecting drug use a Those not included in the categories “ART naïve at TB diagnosis” and”ART use at TB diagnosis” have previously received ART and interrupted their treatment b Those with missing information for chest x-ray (n = 278), HBsAg (n = 215), Anti-HCV (n = 220) assumed negative c Index based on symptoms cough, fever, weight loss d Omitted because of multicollinearity with Treatment Naivety (correlation co-efficient 0.7817) e Omitted in model as part of extra-pulmonary/disseminated In the unadjusted Cox proportional hazard models for all-cause mortality with 24 months follow-up, PLWH with a delayed TB diagnosis were at 36% increased risk of death (hazard ratio 1.36; 95% CI 1.04–1.77; p = 0.023). In the adjusted analysis, the estimate decreases only slightly (adjusted hazard ratio 1.27; 95% CI 0.95–1.70; p = 0.103) (Table 2). At the same time, there is a clear association between increased risk of death and male sex, low or missing CD4 cell count, prior AIDS, disseminated TB disease, weight loss and MDR-TB (Table 2). Stratifying the Cox proportional hazards model for all-cause mor- tality at 0–2 months and 2–24 months follow-up accord- ing to the appearance of the survival curves, no effect of delayed diagnosis on risk of death was observed for the first 2 months (aHR 1.09; 95% CI 0.69–1.71; p = 0.723), while the period 2–24 months seems to explain the Fig. 1 Kaplan–Meier survival estimates by diagnostic delay observed overall effect (aHR 1.36; 95% CI 0.93–1.99; (< 1 month versus ≥ 1 month) p = 0.119) (Additional file 1: Tables S1 and S2). Kraef et al. BMC Infect Dis (2021) 21:1038 Page 6 of 12 Table 2 Cox-regression model for survival in Eastern Europe, crude and adjusted hazard ratios (24 months follow-up) Total (n = 740) Crude HR (95% CI) p-value Adjusted HR (95% CI)a p-value Delayed diagnosis ≥ 1 month (n = 480) 1.36 (1.04–1.77) 0.023 1.27 (0.95–1.70) 0.103 Gender Male (n = 563) 1.34 (1.02–1.76) 0.032 1.38 (1.04–1.83) 0.026 Age Age 18–49 (n = 694) Ref Ref Age ≥ 50 (n = 46) 0.92 (0.55–1.52) 0.741 Exposure group (HIV) MSM (n = 10) 0.49 (0.12–1.95) 0.309 IDU (n = 422) 1.19 (0.93–1.52) 0.173 Heterosexual (n = 183) 0.84 (0.63–1.13) 0.255 Known HIV positive At baseline (n = 668) 1.55 (0.96–2.50) 0.074 1.43 (0.84–2.42) 0.185 Treatment history (HIV) ­ iagnosisc (n = 558) Naïve at TB d 1.04 (0.79–1.38) 0.766 ­ aselinec (n = 132) ART at b 0.96 (0.70–1.31) 0.795 Cotrimoxazole at baseline (n = 273) 1.22 (0.96–1.56) 0.110 CD4-cell count/mm3° 0–199/mm3 (n = 440) 2.17 (1.53–3.09) 0.001 1.84 (1.27–2.68) 0.001 ≥ 200/mm3 (n = 185) Ref Ref Ref Ref Missing CD4 (n = 115) 2.17 (1.41–3.33) 0.001 2.32 (1.47–3.66) 0.001 Prior AIDS Yes (n = 177) 1.87 (1.45–2.41) 0.001 1.69 (1.27–2.24) 0.001 Previous TB Yes (n = 99) 0.97 (0.68–1.39) 0.880 TB risk factor Alcohol abuse (n = 186) 1.32 (1.01–1.73) 0.041 1.24 (0.92–1.68) 0.158 Recent TB in family (n = 53) 0.84 (0.50–1.42) 0.524 Prison within the last 2 years (n = 143) 1.11 (0.82–1.51) 0.493 Clinical presentation of TB Pulmonary (n = 239) Ref Ref Ref Ref Extrapulmonary (n = 54) 0.98 (0.51–1.88) 0.957 0.86 (0.44–1.68) 0.657 Disseminated (n = 447) 2.31 (1.71–3.11) 0.001 1.93 (1.38–2.69) 0.001 TB symptoms at diagnosis Cough (n = 501) 0.95 (0.74–1.23) 0.708 Fever (n = 636) 1.59 (1.07–2.37) 0.022 1.41 (0.91–2.19) 0.124 Weight loss (n = 445) 1.70 (1.31–2.21) 0.001 1.41 (1.05–1.88) 0.022 Number of symptoms§ None (n = 20) Ref Ref One symptom (n = 133) 0.89 (0.38–2.11) 0.793 Two symptoms (n = 312) 1.12 (0.49–2.56) 0.783 Three symptoms (n = 275) 1.58 (0.69–3.58) 0.277 Screening chest X-ray in ­pastd Yes (n = 338) 1.30 (1.02–1.66) 0.034 1.27 (0.93–1.74) 0.133 Rifampicin resistance Yes (n = 95) 1.53 (1.12–2.10) 0.007 Omittedb MDR-TB Yes (n = 88) 2.61 (1.68–4.05) 0.001 2.06 (1.06–4.01) 0.033 No (n = 148) Ref Ref Ref Missing (n = 504) 1.84 (1.28–2.64) 0.001 1.32 (0.68–2.57) 0.407 Diagnosis Definitive (n = 333) Ref Ref Ref Ref Probable (n = 76) 1.43 (0.97–2.11) 0.074 0.99 (0.61–1.59) 0.955 Presumptive (n = 331) 1.03 (0.80–1.34) 0.795 0.95 (0.67–1.35) 0.764 Hepatis B HBsAg positive (n = 44)d 0.95 (0.56–1.60) 0.843 Hepatitis C Anti-HCV positive (n = 405)d 1.20 (0.94–1.53) 0.145 Drug-susceptibility testing At baseline (n = 236) 0.84 (0.65–1.08) 0.175 Treatment with RHZ At baseline (n = 591) 0.81 (0.59–1.09) 0.157 Treatment with at least three active drugs At baseline (n = 132) 0.64 (0.46–0.91) 0.014 0.81 (0.52–1.25) 0.344 HR hazard ratio, CI confidence interval, MSM Men Who Have Sex with Men, IDU injecting drug use Global Test Proportional Hazard Assumption p = 0.3689, individually all > 0.05 a Adjusted for diagnostic delay, gender, known HIV positive, CD4 cell count, prior aids, tb risk factor alcohol, clinical presentation (all), Tb symptoms (fever, weight loss), chest x-ray, MDR-TB, type of diagnosis, treatment with at least three active drugs, stratified by Center b Omitted Rifampicin resistance in the multivariable model due to multicollinearity with MDR-TB c Those not included in the categories “ART naïve at TB diagnosis” and”ART use at TB diagnosis” have previously received ART and interrupted their treatment d Those with missing information for chest x-ray (n = 278), HBsAg (n = 215), Anti-HCV (n = 220) assumed negative § Index based on symptoms cough, fever, weight loss Kraef et al. BMC Infect Dis (2021) 21:1038 Page 7 of 12 Sensitivity analysis better ascertainment of the symptom weight loss and dis- Of all 262 deaths within 24 months, 97 (37.0%) were seminated disease presentation in our study or to uncon- directly TB-related. In an adjusted Cox proportional haz- trolled circularity. ards model for TB-related mortality within 24 months In patients diagnosed with TB (without HIV) in Kiev follow-up delayed diagnosis had no effect on TB-related city long diagnostic delay was reported by individuals death (aHR 1.00; 95% CI 0.62–1.59; p = 0.988). who were homeless, jobless or abused alcohol [24]. In When stratifying delayed diagnosis by ≤ 3 months Uzbekistan TB mono-infected patients self-medication, (n = 640) versus > 3 months (n = 100), at 24 months 247 coughing, loss of weight, and visiting private and pri- (38.6%) deaths had occurred in the group of ≤ 3 months mary healthcare facilities were associated with delay [25]. and 59 (59%) in the group of > 3 months. Kaplan–Maier Similar results were observed in Georgia where receipt survival analysis showed no difference in survival func- of medication prior to TB diagnosis was associated with tion (log-rank p = 0.777) (Additional file 1). increased overall delayed diagnosis [26]. In an adjusted Cox proportional hazards model with Disseminated TB was associated with delayed diagno- 24 months of follow-up there was no effect of a delayed sis in the present analysis. These diagnostic challenges diagnosis for more than three months on all-cause mor- have been described previously in PLWH as the extra- tality (aHR 1.00; 95% CI 0.7–1.45; p = 0.986). pulmonary or disseminated TB presentation makes it dif- ficult to obtain suitable specimen for culture [27, 28]. Discussion Effect of delayed diagnosis on survival in PLWH In the present study, we found older age (≥ 50 years), In our study, those with a delayed diagnosis of at least IDU, ART naivety, disseminated TB and presenting with 1 month had a 36% higher risk of dying within the first weight loss were associated with delayed diagnosis, while 2 years after the TB diagnosis. Compared to previous a previous TB diagnosis was associated with a lower risk studies in other settings the effect seems less pronounced, of late diagnosis. Furthermore, a delayed TB diagnosis and when adjusting for other patient characteristics, the was associated with an increased risk for all-cause death result was no longer insignificant, although the effect size in PLWH in Eastern Europe. Our study provides indica- remains similar and the upper limit of the CI suggests it tion for a systematic screening and active case finding in could be a quite large effect. Furthermore, there is a clear PLWH at risk of delayed TB diagnosis. association between increased risk of death and male sex, low or missing CD4 cell count, prior AIDS, disseminated TB disease, weight loss and MDR-TB; all of which have Factors associated with delayed diagnosis in PLWH been found and described in a previous study of the cur- Our study is the first to present factors associated with rent cohort [17]. In a retrospective cohort of PLWH with delayed TB diagnosis in PLWH in Eastern Europe. Some pulmonary TB in China, a delayed diagnosis of more of the factors identified in the present study, such as older than 1 month from onset of symptoms increased the risk age, being ART naive and IDU have been identified in of death almost three-fold (aHR 2.60; 95% CI 1.42–4.78) other settings. In a South African cohort including only [29]. In a Brazilian study, PLWH with TB who had pre- PLWH with pulmonary TB older age (> 40 years) and viously started on ART, duration of TB symptoms over HIV-RNA viral load > 400 copies/ml were predictors for 3 months before diagnosis increased the risk of death delayed diagnosis, while being on ART at time of diag- six-fold (HR = 6.15, 95% CI 1.15–32.9) [14]. In Malawi, nosis was associated with lower risk of delayed diagno- seeking care after more than 1 month or 3 months was sis [20]. A study including all drug-susceptible TB cases associated with an increased risk of death of about 60% in PLWH from Colombia found age between 15–34, (aHR 1.6, 95% CI 0.9–2.8) and 100% (aHR 2.0, 95% CI age ≥ 45 years, and receiving tuberculin skin tests as a 1.1–3.8), respectively [30]. On the contrary, a Thai study part of the diagnostic work-up associated with delayed of 667 PLWH with TB found no effect of delayed TB diagnosis [21]. In a Brazilian cohort of 242 PLWH with diagnosis on survival [31]. TB, factors associated with patient delay were the symp- Supplementary survival analyses stratified by follow- toms asthenia, chest pain, and the use of illicit drugs, up time suggest that delayed diagnosis has no effect on including IDU [22]. In this current study, we identi- survival during the first 2 months after TB treatment fied a previous TB diagnosis as associated with lower initiation, but seem to have an effect beyond 2 months risk of delayed diagnosis, which is in line with another after initiation of TB treatment, albeit insignificantly cohort of PLWH from Brazil [23]. Weight loss and dis- so, perhaps due to limited number of patients in the seminated disease have not been previously described as study. In our study, mortality was very high during the associated with delayed diagnosis which could be due to first 2 months, which leads us to speculate that those Kraef et al. BMC Infect Dis (2021) 21:1038 Page 8 of 12 presenting with shorter symptom duration might be suf- an active case finding strategy [38]. Periodic screening fering from quickly progressing disease with worse early for TB among PLWH using symptoms, chest x-ray, Xpert outcomes. That suggests that other factors were driving and/or urine-LAM point-of-care testing according to the mortality in the first 2 months diluting the effect of WHO guidelines could improve case finding [39]. How- delayed diagnosis on survival in the early period. These ever, to the best of our knowledge no evidence on the could be factors such as disseminated disease, MDR-TB systematic implementation of these guidelines in Eastern and an AIDS diagnosis which also are significant predic- Europe and Central Asia exists. tors of mortality in the stratified analysis 0–2 months Our study has some limitations. First, the generalizabil- (Additional file 1: Table S1). ity of our results to the background/general population may be impaired due to selection bias, as recruitment Potential implications for policy and research to our study is conditional on both TB and HIV being It is very likely that better integration of clinical services diagnosed. Those who were never diagnosed with TB or and diagnostic capacity can decrease delays in diagnos- HIV due to early death without making contact to health ing TB in PLWH. HIV and TB services in Eastern Europe care were not included. However, these patients presum- are only provided in one place at about 36% of health care ably would have worse outcomes and a long diagnostic centers [32]. While evidence of the effect of integrated delay if included here, as reduced access to health care health care on delayed diagnosis is not available, studies is likely associated with similar risk factors as delayed in Sub-Saharan Africa have shown improved HIV and TB diagnosis, morbidity and mortality. Therefore, our results treatment outcomes when integrating HIV and TB care likely underestimate the effect of delayed diagnoses on [33, 34]. mortality in the population. Secondly, Eastern Europe is Furthermore, strengthening diagnostic capacity for heterogeneous, delay of diagnosis may vary within the sputum-smear negative and extra-pulmonary TB (i.e., region and in some countries, health care has improved molecular diagnostic tests such as Xpert) are crucial as since the period of data collection in 2013 [32]. As drug TB in PLWH often presents as extrapulmonary or dis- susceptibility testing was not widely available in East- seminated TB and is commonly sputum-smear nega- ern Europe at the time of the study and those without tive. There is evidence from Mozambique, where PLWH testing results were assumed susceptible our study very delayed for TB diagnosis due to initial negative sputum likely underestimated MDR-TB and has led to increased smear microscopy but consecutive positivity (Xpert TB mortality. Thirdly, we cannot rule out a certain degree of test) had a 12-fold higher mortality, reiterating the impor- circularity in the analysis of risk factors for delayed diag- tance of rapid diagnostics [16]. In Rwanda, expanded nosis; e.g., disseminated disease or weight loss can be risk access to systematic Xpert testing and standardized treat- factors for delayed diagnosis due to an unexpected or ment of Rifampicin-resistant TB in both HIV-seroneg- hard to diagnosis disease presentation but they can also ative and PLWH led to the reduction of diagnostic and be a consequence of delayed diagnosis. Finally, for most treatment delays resulting in reduced mortality [35]. In patients in the present study the date of TB diagnosis and Eastern Europe, Xpert was used as part of routine work- of treatment initiation were identical. Therefore, we were up at almost all (92%) clinics in 2018, which is a major not able to calculate the health care delay, the time span improvement compared to 2013 (54%) when the present between diagnosis and treatment initiation. study was conducted, so we cannot rule out improve- Regardless, the current study adds important knowl- ments to delayed diagnosis in the meantime [32]. A edge to the field of health care for TB/HIV co-infection promising additional tool for rapid diagnosis and treat- in Eastern Europe. ment initiation at the point of care for PLWH regardless of site of TB infection is the urine-based lipoarabinoman- Conclusions nan (LAM) test [36]. Further prospective studies of this We found that nearly two thirds of PLWH with TB in intervention and its effect on delayed diagnosis, in par- Eastern Europe had a delayed TB diagnosis, in particular ticular in an Eastern European context, are needed. those of older age, people who inject drugs, ART naïve, In addition to improving health care infrastructure, with disseminated disease, and presenting with weight active case finding among high-risk populations, in our loss. Patients with delayed TB diagnosis were subse- study those of older age, people who inject drugs and quently at higher risk of death in unadjusted analysis. PLWH not on ART, may help reduce delayed diagnoses There is a clear need for optimization of the current TB and improve linkage to health care [37]. In an active pul- diagnostic cascade and HIV care in PLWH in Eastern monary TB case finding study among newly HIV diag- Europe to reduce delayed diagnoses to reduce morbid- nosed patients in Tbilisi, Georgia, 11.5% of patients were ity and improve treatment outcomes. While health care identified with pulmonary TB, highlighting the utility of planners and policy makers should ensure strengthened Kraef et al. BMC Infect Dis (2021) 21:1038 Page 9 of 12 diagnostic capacity, clinicians should pay special atten- ­ rabtree78, JL Mosqueda ­Gomez79, J. A. ­Villanueva80; LA C tion to patient groups identified as being at higher risk Gonzalez ­Hernandez80 and F. ­Badial80. for delayed TB diagnosis, and further research should 14 Belarusian State Medical University, Department of focus on developing more effective diagnostic interven- Infectious Disease. tions to reduce delayed diagnoses. 15 Republican Research and Practical Centre for Pulmonology. 16 Gomel State Medical University (Gomel). 17 Gomel Region Centre for Hygiene: O. Suetnov (PI) Appendix TB:HIV Study Group and D. Paduto. w w w . ​ c h i p . ​ d k / ​ R e s e a ​ r c h / ​ S t u d i ​ e s / ​ T B H I V /​ 18 East Viru Central Hospital (Kohtla-Jarve), Estonia. TBHIV-​Study-​Group 19 Infectious Diseases, AIDS and Clinical Immunology The Study Group consists of the following 62 partici- Research Center (Tiblisi). pating sites in 19 countries (listed with countries, names 20 National Center for Tuberculosis and Lung Diseases of participating centers). of Georgia (Tibilisi). A. ­Vassilenko14, D. K ­limuk15, O. K ­ondratenko15, 21 Infectology Centre of Latvia (Riga). 15 16 A. ­Zalutskaya , V. ­ Bondarenko , V. ­ Mitsura16, E. 22 Centre for Communicable Diseases and AIDS 16 16 17 ­Kozorez , O. T ­ umash , O. S ­ uetnov , D. P ­ aduto17, V. (Vilnius). 18 18 19 ­Iljina , T. K ­ ummik , K. M ­ shvidobadze , N. L ­ anchava19, 23 Wojewodski Szpital Zakanzy/Medical University of 20 20 L. ­Goginashvili , L. M ­ ikiashvili , N. ­ Bablishvili20, Warsaw (Warszawa). 21 21 B. ­Rozentale , I. ­ Zeltina , I. ­ Janushkevich21, I. 24 Wojewodski Szpital Specjalistyczny/Medical Uni- 22 22 ­Caplinskiene , S. ­ Caplinskas , Z. ­ Kancauskiene22, versity Teaching Hospital (Bialystok). 23 A. Wiercinska-Drapalo , M. ­ompson24, Th J. 25 Jozef Strus Multidisciplinary City Hospital (Poznan). 24 24 ­Kozlowska , A. ­ Grezesczuk , M. B ­ura25 B. ­ Knysz26, 26 Wroclaw University School of Medicine (Wroclaw). M. ­Inglot26, A. G ­ arlicki27, J. L ­ oster27, D. D ­ uiculescu28, 27 Jagiellonian University Medical College (Krakow). 29 29 A. ­Rakhmanova , O. P ­anteleev , A. ­ Yakovlev29, A. 28 Dr. Victor Babes Hospital (Bucharest). 29 29 29, 30 ­Kozlov , A. T ­ yukalova , Y. V ­ lasova , T. T ­ rofimov31, 29 Botkin Hospital of Infectious Diseases (St. 32 33 33 G. ­Kyselyova , N. O ­ bel , J. G ­ erstoft , G. K ­ ronborg34, Petersburg). 35 35 35 MC ­Payen , K. ­ Kabeya , C. ­ Necsoi , F. ­ Dabis36, A. 30 City TB Hospital No. 2 (St. Petersburg): ­Tsaranazy , C. ­Cazanave , H. ­Furrer , M. ­Sagette38, M. 36 37 38 31 Center for Prevention and Control of AIDS (Veliky, ­Rickenbach38, D. ­Sculier39, A. ­Calmy39, M. C ­ avassini40, A. Novgorod). 41 41 42 ­Bruno , E. ­Bernasconi , M. H ­ offmann , P. V ­ ernazza42, 32 Crimean Republican AIDS Centre (Simferopol). J. ­Fehr , R. ­Weber , R. ­Miller , N. ­Vora , G. ­Cooke45, 43 43 44 44 33 Rigshospitalet (Cph). S. ­Mullaney45, E. ­ Wilkins46, V. ­ George46, P. ­ Collini47, 34 Hvidovre University Hospital. 47 48 48 D. ­Dockrell , F. P ­ost , L. C ­ ampbell , R. B ­ rum48, E. 35 CHU Saint-Pierre (Brussels). 48 48 49 ­Mabonga , P. S ­aigal , S. K ­ egg , J. A ­ insworth50, A. 36 Aquitaine Cohort. 50 51 51 52 ­Waters J. D ­ har , K. E ­ llis , E. G ­ irardi , A R ­ ianda52, V. 37 Bordeaux University Hospital. 52 52 52 ­Galati , C. P ­ innetti , C. T ­ ommasi , G. L ­ apadula53, A. 38 University Hospital Bern. Di ­Biagio , A. ­Parisini , S. ­Carbonara , G. ­Angarano55, 54 54 55 39 Hopital Cantonal Universitaire, Geneve. M. ­Purgatorio55, A. M ­ atteelli56, A. A ­postoli56, JM. 40 Centre Hospitalaire Universitaire Vaudois, ­Miro , C. ­Manzardo , C. ­Ligero , J. ­Gonzalez57, Jose 57 57 57 Lausanne. A. Martinez-Martinez57, F. ­Sanchez58, H. ­Knobel58, M. 41 Hospital of Lugano. Salvadó58, J.L. Lopez-Colomes58, X. Martínez-Lacasa59, 42 Cantonal Hospital St. Gallen. E. Cuchí59, V. Falcó60, A. ­ Curran60, M.T. ­ Tortola60, I. 43 University Hospital Zurich. Ocaña , R. ­Vidal , MA. ­Sambeat , V. ­Pomar61, P. ­Coll61, 60 60 61 44 Mortimer Market Centre (London). D. ­Pozamczer62, M. ­Saumoy62, F. ­Alcaide62, J. Caylà63, 45 St. Mary’s Hospital. A. ­Moreno63, J.P. ­ Millet63, A. ­ Orcau63, L. ­ Fina63, A. 46 North Manchester General Hospital. ­Romero , L.L. ­Roldan , JA. ­Iribarren , M. ­Ibarguren64, 63 63 64 47 Sheffield Teaching Hospitals. S. ­Moreno65, A. González65, P. M ­ iralles66, T. Aldámiz- 48 King’s College Hospital (London). 66 67 Echevarría , M. ­Losso , J. T ­ oibaro67, L. ­Gambardella67, J. 49 Queen Elizabeth Hospital. ­Toibaro , L. Moreno ­Macias68, E. ­Warley69, S. ­Tavella69, 68 50 North Middelsex University Hospital. O. Garcia M ­ essina70, O. G ­ear70, H. L ­aplume71, C. 51 Leicester Royal Infirmary. 72 73 ­Marson , J. ­Contarelia , M. M ­ ichaan , P. ­Scapellato74, 73 52 IRCCS—Ospedale L. Spallanzani (Rome). B. ­Bartoletti , D. ­Palmero , C. ­Elias76, C. ­Cortes77, B. 75 75 53 AO San Gerardo (Monza). Kraef et al. BMC Infect Dis (2021) 21:1038 Page 10 of 12 54 IRCCS AOU San Martino – IST di Genoa Authors’ contributions CK, OK and AM conceived of the study. AP, IA, AS, NB, ST, RP, IK, EB, ED, JL, (Genova). DP contributed to data collection. CK did statistical analysis. CK, AB, AM, OK, 55 Clinic of Infectious Diseases, University of Bari DP and ISJ contributed to interpretation of data analysis. All authors were (Bari). involved in drafting the final manuscript. All authors read and approved the final manuscript. 56 University of Brescia Spedali Civili. 57 Hospital Clinic of Barcelona. Funding 58 Hospital del Mar. Open Access funding enabled and organized by Projekt DEAL. 59 Mutua de Terrassa. Availability of data and materials 60 Hospital Universitari Vall d’Hebrón: The datasets used and/or analyzed during the current study are available from 61 Hospital Universitari de la Santa Creu i Sant Pau. the corresponding author on reasonable request. 62 Hospital Universitari de Bellvitge. 63 Agencia de Salud Pública de Barcelona. Declarations 64 Hospital Universitaria Donostia (San Sebastian). Ethics approval and consent to participate 65 Hospital Ramon y Cajal (Madrid). Informed consent was obtained from all participants. All methods were car- 66 Hospital Universitaria’Gregorio Marañon’’ (Madrid). ried out in accordance with relevant guidelines and regulations. In specific, IEC/IRB were obtained from the following commitees: 67 The CICAL Cohort, Argentina. Le Comité d’Ethique du C.H.U Saint-Pierre, Le Numéra registre AK/12-03- 68 Hospital J. M. Ramos Mejía (Buenos Aires). 28/4128, Comité. 69 Hospital Paroissien (BA). Ético Cientifico del servicio de Salud Metropolitano Centra, Chil, certificado 452/11, De. 70 Hospital Piñero (BA). Videnskabsetiske Komiteer i Region Hovedstaden Journal nr.: H-3-2011-095, 71 Hospital Nacional Profesor Alejandro Posadas: United Kingdom national ethics approval Reference nr. 11/LO/0713 & R&D 72 Hospital Rawson (Cordoba). Reference nr. CSP 75430 for UK sites: Mortimor Market Centre, London, Impe- rial College Healthcare, London, St. Marys Hospital, London, North Manchester 73 Hospital San Juan de Dios (La Plata). General Hospital, Manchester, Sheffield Teaching Hospitals, Sheffield, King’s 74 Hospital General de Agudos Donación F. Santojani. College Hospital, London, North Middlesex University Hospital, London, 75 Hospital Francisco Javier Muñiz (BA). Queen Elisabeth Hospital, London R&D ref. nr. SLHT/2011/UCSM/HIV/88, Leicester Royal Infirmary, Leicester. El Comité de Etica en Investigación en 76 Hospital Jujuy. Salud, Hospital Paroissien Argentina 20/12/2911, Comité de Etica Iniciativa 77 Fundación Arriaran (Santiago). y Reflextion Bioetica Hospital Pineiro & Hospital Ramos Mejia Argentina 78 INNcMZS (México DF). 12/12/12, Comité de Bioética Reg. nr. 11,743/11 Hospital Nacional Profesor Alejandro Posdas Argentina, Comité Institucional de Ètica de la Investigacion 79 Hospital General Regional de Leon- CAPACITS. en Salud del Nino u del Adulto Cordoba 16/12/2012 Hospital Rawson, Comité 80 Hospital Civil de Guadalajara. de Etica La Plata 12/11/12 Hospital Sa Juan de Dios Argentina, Comité de Ética en Investigacion Buenos Aires 18/11/12 Hospital Santojani Argentina & 11/05/2012 Hospital Fracisco Javier Muniz Argentina, Docencia e Investi- Abbreviations gación 15/06/12 Jujuy Argentina, Institutional Review Board Tbilisi Ref. nr. TB: Tuberculosis; HIV: Human immunodeficiency virus; PLWH: People living 12–007, Registro delle Sperimentazioni del Comitato Etico Rone nr. 24/2011 with HIV; AIDS: Acquired immunodeficiency syndrome; IEC: Independent Spallanzani, Comitato Etico San Gerardo, Monza 26/05/2011, Comitato Etico Ethics Committee; IRB: Institutional Review Board; RNA: Ribonucleotide acid; Aziendale A.O.U. San Martino Genoa nr. 295 08/04/2011, Comitato Etico ART​: Antiretroviral therapy; HBsAg: Hepatitis B surface antigen; HCV: Hepatitis Indipendente Locale Policlinio Consorziale Bari nr. 624 28/09/2011, Instituto C virus; MDR: Multi-drug resistant; PCR: Polymerase chain reaction; Mtb: Nacional de Ciencias Médicas y Nutrición Salvador Zubirán Ref. nr 437, Comité Mycobacterium tuberculosis; CSF: Cerebrospinal fluid; KM: Kaplan–Meier; HR: de Ética del Hospital General Regional de León 10/10/2011, Secretario del Hazard ratio; CI: Confidence interval; IDU: Injecting drug use; OR: Odds ratio; Comité de Ensenanza, Investigacion y Ética Hospital Civil de Guadalajara PYFU: Person years of follow-up; LAM: Lipoarabinomannan. Invest. Nr. 038/12, Comitato Etico Spedali Civili Brescia 07/06/2011 nr.36; Direc- ció General de Regularcio Barcelona CY-ANT-2011-01, Hospital Universitario Ramón y Cajal Comité Ético de Investigación Clinica 12/07/2011, Comite Supplementary Information Etico de Investigation Clinica de Euscadi Vitoria 02/02/2012, Comité Ético de The online version contains supplementary material available at https://​doi.​ Investigación Clínica Hospital General Universitario Gregorio Maranon Madrid org/​10.​1186/​s12879-​021-​06745-w. SAS/3470/2009, Kantonale Ethik-Kommission Zürich EK-793, Ethical Com- mittee of Rep. Res. And Practical Centre for Pulmonology Minsk 11/02/2011, Ethical Committee of Gomel State Medical University & Gomel Regional Additional file 1: Table S1. Cox-regression model for survival in Eastern Centre for Hygiene, Ethical Committee for Botkin Hospital & City TB Hospital Europe, crude and adjusted hazard ratios for follow-up 0–2 months. ref. nr.23, Ethical Committee for Novgorod Centre for AIDS Prevention and Table S2. Cox-regression model for survival in Eastern Europe, crude Control 11/10/2011, Ethican Committee of Samara State Medical University and adjusted hazard ratios for follow-up 2–24 months. Figure S1. 10/10/2012, Komisja Bioetycznego Uniwersytetu Medycznego w Bialymstoku Kaplan–Meier survival estimates by diagnostic delay (≤ 3 months versus Poland R-I-00/85c/2012 –R-I-002/85/2011, Tallinn Medical Research Ethics > 3 months) Committee Appr. Nr. 2555, RSU Etikas Komitejas Lemums Riga Nr. E-9(2), Vilniaus Regioninis Biomedicininiu Tyrimu Etikos Komitetas Nr. 158,200–07– 363-90, Dr. Victor Babes Hospital Institutional Ethics Committee nr. 3889, Ethics Acknowledgements Committee Crimean Republican AIDS Centre Nr. 462–13–08-12. Not applicable. The data storage and handling was protected in accordance with and approved by the Danish Data Protection Agency (Datatilsynet) under the Act Guidelines on Processing of Personal Data (Act No. 429 of 31 May 2000) and European The STROBE guidelines for cohort studies were followed in the conduct of this Commission Directive 95/46/EC. http://​www.​chip.​dk/​Porta​ls/0/​files/​Study%​ study. 20doc​uments/​HIV_​TB_​proje​ct_​proto​col_​2011.​pdf. Kraef et al. BMC Infect Dis (2021) 21:1038 Page 11 of 12 Consent for publication 12. Lin X, Chongsuvivatwong V, Lin L, Geater A, Lijuan R. Dose-response Not applicable. relationship between treatment delay of smear-positive tuberculosis patients and intra-household transmission: a cross-sectional study. Trans Competing interests R Soc Trop Med Hyg. 2008;102(8):797–804. AM reports consulting fees by ViiV, Gilead, Eiland and Bonnin PC outside of 13. WHO | WHO end TB strategy. WHO. World Health Organization. http://​ the submitted work. RP reports fees for lectures and travel by Gilead sciences www.​who.​int/​tb/​post2​015_​strat​egy/​en/. Accessed 5 Aug 2020. Poland outside of the submitted work. OK reports fees for consultation and 14. Schmaltz CAS, Santoro-Lopes G, Lourenço MC, Morgado MG, Velasque travel by Gilead, Janssen, Merck and Viiv outside the submitted work. All other LdS, Rolla VC. Factors impacting early mortality in tuberculosis/HIV authors declare that they have no conflict of interest. patients: differences between subjects naïve to and previously started on HAART. PLoS ONE. 2012;7(9):e45704. Author details 15. Finnie RKC, Khoza LB, van den Borne B, Mabunda T, Abotchie P, Mullen 1 CHIP (Centre of Excellence for Health, Immunity and Infections), Rigshospita- PD. Factors associated with patient and health care system delay in diag- let, University of Copenhagen, Copenhagen, Denmark. 2 Department of Infec- nosis and treatment for TB in sub-Saharan African countries with high tious Diseases, The Heart Centre, Rigshospitalet, University of Copenhagen, burdens of TB and HIV. Trop Med Int Health. 2011;16(4):394–411. Copenhagen, Denmark. 3 City TB Dispensary, St. Petersburg, Russia. 4 Republi- 16. Lisboa M, Fronteira I, Colove E, Nhamonga M, Martins MdRO. Time delay can Scientific and Practical Center for Pulmonology and Tuberculosis, Minsk, and associated mortality from negative smear to positive Xpert MTB/ Belarus. 5 Infectious Diseases, AIDS and Clinical Immunology Research Center, RIF test among TB/HIV patients: a retrospective study. BMC Infect Dis. Tbilisi, Georgia. 6 Dr Victor Babes’ Hospital of Tropical and Infectious Diseases, 2019;19(1):18. Bucharest AND ‘Carol Davila’ University of Medicine and Pharmacy, Bucharest, 17. Podlekareva DN, Efsen AMW, Schultze A, Post FA, Skrahina AM, Panteleev Romania. 7 Wojewodski Szpital Zakanzy/Medical University of Warsaw, Warsaw, A, et al. Tuberculosis-related mortality in people living with HIV in Poland. 8 Department of Infectious Disease, Belarusian State Medical University, Europe and Latin America: an international cohort study. Lancet HIV. Minsk, Belarus. 9 Samara State Medical University of the Ministry of Healthcare 2016;3(3):e120–31. of the Russian Federation, Samara, Russia. 10 Botkin Hospital of Infectious 18. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke Disease, St. Petersburg, Russia. 11 Riga East University Hospital, Latvian Centre JP, et al. The Strengthening the Reporting of Observational Studies in of Infectious Diseases, Riga, Latvia. 12 Research Unit for Infectious Diseases, Epidemiology (STROBE) statement: guidelines for reporting observational Odense University Hospital, University of Southern, Odense, Denmark. 13 Cen- studies. Lancet. 2007;370(9596):1453–7. tre for Clinical Research, Epidemiology, Modelling and Evaluation (CREME), 19. Efsen AMW, Schultze A, Miller RF, Panteleev A, Skrahin A, Podlekareva Institute for Global Health, UCL, London, UK. 14 Heidelberg Institute of Global DN, et al. Management of MDR-TB in HIV co-infected patients in Eastern Health, University of Heidelberg, Heidelberg, Germany. Europe: results from the TB:HIV study. J Infect. 2018;76(1):44–54. 20. Boniface R, Moshabela M, Zulliger R, MacPherson P, Nyasulu P. Correlates Received: 13 April 2021 Accepted: 20 August 2021 of delayed diagnosis among human immunodeficiency virus-infected pulmonary tuberculosis suspects in a rural HIV clinic, South Africa. Tuberc Res Treat. 2012;2012. 21. Ríos-Hincapié CY, Rojas M, López M, Porras A, Luque R, Pelissari DM, et al. Delays in HIV and TB diagnosis and treatment initiation in co-infected References patients in Colombia. Int J STD AIDS. 2020;31(5):410–9. 1. World Health Organization. Global tuberculosis report 2020. Geneva: 22. Coimbra I, Maruza M, Militão-Albuquerque MdFP, Moura LV, Diniz GTN, World Health Organization; 2020. Miranda-Filho DdB, et al. Associated factors for treatment delay in pulmo- 2. Ford N, Matteelli A, Shubber Z, Hermans S, Meintjes G, Grinsztejn B, et al. nary tuberculosis in HIV-infected individuals: a nested case-control study. TB as a cause of hospitalization and in-hospital mortality among people BMC Infect Dis. 2012;12:208. living with HIV worldwide: a systematic review and meta-analysis. J Int 23. Nogueira BMF, Rolla VC, Akrami KM, Kiene SM. Factors associated with AIDS Soc. 2016;19(1):20714. tuberculosis treatment delay in patients co-infected with HIV in a high 3. Gupta RK, Lucas SB, Fielding KL, Lawn SD. Prevalence of tuberculo- prevalence area in Brazil. PLoS ONE. 2018;13(4):e0195409. sis in post-mortem studies of HIV-infected adults and children in 24. van der Werf MJ, Chechulin Y, Yegorova OB, Marcinuk T, Stopolyanskiy A, resource-limited settings: a systematic review and meta-analysis. AIDS. Voloschuk V, et al. Health care seeking behaviour for tuberculosis symp- 2015;29(15):1987–2002. toms in Kiev City, Ukraine. Int J Tuberc Lung Dis. 2006;10(4):390–5. 4. Getnet F, Demissie M, Assefa N, Mengistie B, Worku A. Delay in diagnosis 25. Belkina TV, Khojiev DS, Tillyashaykhov MN, Tigay ZN, Kudenov MU, of pulmonary tuberculosis in low-and middle-income settings: system- Tebbens JD, et al. Delay in the diagnosis and treatment of pulmonary atic review and meta-analysis. BMC Pulm Med. 2017;17:202. tuberculosis in Uzbekistan: a cross-sectional study. BMC Infect Dis. 5. Storla DG, Yimer S, Bjune GA. A systematic review of delay in the diagno- 2014;25(14):624. sis and treatment of tuberculosis. BMC Public Health. 2008;14(8):15. 26. Rabin AS, Kuchukhidze G, Sanikidze E, Kempker RR, Blumberg HM. Pre- 6. Tedla K, Medhin G, Berhe G, Mulugeta A, Berhe N. Delay in treatment ini- scribed and self-medication use increase delays in diagnosis of tubercu- tiation and its association with clinical severity and infectiousness among losis in the country of Georgia. Int J Tuberc Lung Dis. 2013;17(2):214–20. new adult pulmonary tuberculosis patients in Tigray, northern Ethiopia. 27. Whitehorn J, Ayles H, Godfrey-Faussett P. Extra-pulmonary and smear- BMC Infect Dis. 2020;20(1):456. negative forms of tuberculosis are associated with treatment delay and 7. Bajehson M, Musa BM, Gidado M, Nsa B, Sani U, Habibu AT, et al. Deter- hospitalisation. Int J Tuberc Lung Dis. 2010;14(6):741–4. minants of mortality among patients with drug-resistant tuberculosis in 28. Hudson CP, Wood R, Maartens G. Diagnosing HIV-associated tuber- northern Nigeria. PLoS ONE. 2019;14(11):e0225165. culosis: reducing costs and diagnostic delay. Int J Tuberc Lung Dis. 8. Jørstad MD, Aẞmus J, Marijani M, Sviland L, Mustafa T. Diagnostic delay 2000;4(3):240–5. in extrapulmonary tuberculosis and impact on patient morbidity: a study 29. Ji Y-J, Liang P-P, Shen J-Y, Sun J-J, Yang J-Y, Chen J, et al. Risk factors affect- from Zanzibar. PLoS ONE. 2018;13(9):e0203593. ing the mortality of HIV-infected patients with pulmonary tuberculosis 9. Lin Y, Enarson DA, Du J, Dlodlo RA, Chiang C-Y, Rusen ID. Risk factors for in the cART era: a retrospective cohort study in China. Infect Dis Poverty. unfavourable treatment outcome among new smear-positive pulmonary 2018;7(1):25. tuberculosis cases in China. Public Health Action. 2017;7(4):299–303. 30. Delay in seeking care for tuberculosis symptoms among adults newly 10. Virenfeldt J, Rudolf F, Camara C, Furtado A, Gomes V, Aaby P, et al. Treat- diagnosed with HIV in rural Malawi. https://​pubmed.​ncbi.​nlm.​nih.​gov/​ ment delay affects clinical severity of tuberculosis: a longitudinal cohort 29471​905/. Accessed 1 Feb 2021. study. BMJ Open. 2014;4(6):e004818. 31. HIV care and treatment factors associated with improved survival during 11. Golub JE, Bur S, Cronin WA, Gange S, Baruch N, Comstock GW, et al. TB treatment in Thailand: an observational study | BMC Infectious Dis- Delayed tuberculosis diagnosis and tuberculosis transmission. Int J eases | Full Text. https://​doi.​org/​10.​1186/​1471-​2334-9-​42. Accessed 2 Feb Tuberc Lung Dis. 2006;10(1):24–30. 2021. Kraef et al. BMC Infect Dis (2021) 21:1038 Page 12 of 12 32. Bentzon AK, Panteleev A, Mitsura V, Borodulina E, Skrahina A, Denisova 37. WHO | Latent TB Infection: updated and consolidated guidelines for E, et al. Healthcare delivery for HIV-positive people with tuberculosis in programmatic management. WHO. World Health Organization. http://​ Europe. HIV Med. 2021;22:283–93. www.​who.​int/​tb/​publi​catio​ns/​2018/​latent-​tuber​culos​is-​infec​tion/​en/. 33. Herce ME, Morse J, Luhanga D, Harris J, Smith HJ, Besa S, et al. Integrat- Accessed 29 May 2021. ing HIV care and treatment into tuberculosis clinics in Lusaka, Zambia: 38. Kempker RR, Chkhartishvili N, Kinkladze I, Schechter MC, Harrington K, results from a before-after quasi-experimental study. BMC Infect Dis. Rukhadze N, et al. High yield of active tuberculosis case finding among 2018;18(1):536. HIV-infected patients using Xpert MTB/RIF testing. Open Forum Infect 34. Owiti P, Zachariah R, Bissell K, Kumar AMV, Diero L, Carter EJ, et al. Dis. 2019;6(6):ofz233. Integrating tuberculosis and HIV services in rural Kenya: uptake and 39. WHO consolidated guidelines on tuberculosis Module 2: screening—sys- outcomes. Public Health Action. 2015;5(1):36–44. tematic screening for tuberculosis disease. https://​www.​who.​int/​publi​ 35. Ngabonziza J-CS, Habimana YM, Decroo T, Migambi P, Dushime A, catio​ns/i/​item/​97892​40022​676. Accessed 8 Apr 2021. Mazarati JB, et al. Reduction of diagnostic and treatment delays reduces rifampicin-resistant tuberculosis mortality in Rwanda. Int J Tuberc Lung Dis. 2020;24(3):329–39. Publisher’s Note 36. Bulterys MA, Wagner B, Redard-Jacot M, Suresh A, Pollock NR, Moreau Springer Nature remains neutral with regard to jurisdictional claims in pub- E, et al. Point-of-care urine LAM tests for tuberculosis diagnosis: a status lished maps and institutional affiliations. update. J Clin Med. 2019;9(1):111. Ready to submit your research ? Choose BMC and benefit from: • fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations • maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions