AEGiS-Glasgow [1758-2652-11-S1-O39] FACTORS ASSOCIATED WITH POOR CLINICAL OUTCOME AMONG HIV-INFECTED PATIENTS WITH TUBERCULOSIS (TB) IN EUROPE AND ARGENTINA. THE HIV/TB COLLABORATIVE STUDY

Ninth International Congress on Drug Therapy in HIV Infection

Glasgow, UK - 9-13 November 2008

FACTORS ASSOCIATED WITH POOR CLINICAL OUTCOME AMONG HIV-INFECTED PATIENTS WITH TUBERCULOSIS (TB) IN EUROPE AND ARGENTINA. THE HIV/TB COLLABORATIVE STUDY

J Int AIDS Soc 2008, 11(Suppl 1):39 doi:10.1186/1758-2652-11-S1-O39

D Podlekareva¹, A Mocroft², FA Post³, V Riekstina⁴, JM Miro⁵, H Furrer⁶, M Bruyand⁷, A Panteleev⁸, E Girardi⁹, JJ Toibaro¹⁰, J Caylá ¹¹, R Miller¹², N Obel¹³, A Skrahin¹⁴, E Malashenkov¹⁵, JD Lundgren¹⁶, O Kirk¹⁶ and HIV/TB study group¹⁶
¹ Copenhagen HIV Programme, University of Copenhagen, Copenhagen, Denmark ² Royal Free and University College Medical School, London, UK ³ King's College London School of Medicine, London, UK ⁴ State Agency of TB and Lung Diseases, Riga, Latvia ⁵ Hosp. Clinic – IDIBAPS, University of Barcelona, Barcelona, Spain ⁶ University Hospital of Bern, Bern, Switzerland ⁷ INSERM, U 897, "epidemiology and biostatistics", Bordeaux, France ⁸ TB Hospital #2, St. Petersburg, Russian Federation ⁹ Instituto Nazionale Malattie Infettive L Spallanzani, Rome, Italy ¹⁰ Hospital JM Ramos Mejia, Buenos Aires, Argentina ¹¹ Barcelona TB Research Unit, Barcelona, Spain ¹² Mortimer Market Centre, London, UK ¹³Dept of Infectious Diseases; Rigshospitalet, Copenhagen, Denmark ¹⁴ Research Institute of Pulmonology and Pulmonary Tuberculosis, Minsk, Belarus ¹⁵ Botkin Hospital of Infectious Diseases, St. Petersburg, Russian Federation ¹⁶ University of Copenhagen, Copenhagen HIV Programme, Faculty of Health Science, Copenhagen, Denmark

PURPOSE OF THE STUDY: TB is a common and potentially fatal co-infection among HIV-infected patients worldwide. We aimed to evaluate potential regional differences in patient characteristics and clinical management and their influence on the oneyear mortality rate after a TB diagnosis in HIV-infected patients across Europe and Argentina.

METHODS: 1,075 consecutive HIV-patients who started treatment for TB between January 2004 and December 2006 in 47 clinics across Europe and Argentina were identified. Patients were stratified according to region of residence: Argentina (A), Southern Europe (S), Central/Northern Europe (CN), or Eastern Europe (E). Deaths among HIV/TB co-infected patients within 12 months of TB diagnosis, and factors associated with death, were analysed.

Figure 1 (abstract O39)

SUMMARY OF RESULTS: At TB diagnosis, there were profound differences in patient characteristics, usage of anti-TB and combination antiretroviral therapy (cART), and anti-TB drug resistance in E compared with the other regions (Table 1). Significantly fewer patients in E initiated cARTwithin the first year after TB diagnosis (Table 1), and multi-drug resistant TB was more common in E (12% [31 of 252 patients with data on anti-TB resistance]) compared to A, S and CN (3%, 2% and 3% respectively, p=0.0002). Progression to death was significantly faster in E compared to other regions (Figure 1). In multivariable Cox models, the adjusted relative hazard of death (RH, compared with E) was 0.44 (95% CI 0.22–0.88), 0.33 (0.17–0.66), 0.46 (0.20–1.05) in A, S and CN, respectively. Other factors significantly associated with increased mortality were: CD4 count < 200 cells/mm³ vs. >200 cells/mm³ [2.27 (1.52–3.40)], prior AIDS vs. no AIDS [1.84 (1.29–2.62)], and disseminated TB vs. not disseminated TB [2.01 (1.14–3.56)]. Patients who started anti-TB treatment with at least four first-line drugs had a significantly lower risk of death [0.50 (0.31–0.81)], as did patients with no resistance to anti-TB drugs [0.48 (0.28–0.79)].

Table 1 (abstract O39)

	A (n=115)	S (n=210)	CN (n=168)	E (n=582)	P-value
Caucasians (%)	23	56	33	83	<0.0001
Injecting drug use (%)	37	35	14	80	<0.0001
>4 1st line anti-TB drugs in initial regimen (%)	83	63	77	25	<0.0001
>1 2nd line anti-TB drug in initial regimen (%)	12	15	10	64	<0.0001
Resistance to any anti-TB drug (%, 513 tests)	7	13	7	50	<0.0001
CD4 count at TB diagnosis (cells/mm³, median, inter-quartile range)	92 (41–228)	146 (55–291)	145 (54–284)	212 (89–463)	<0.0001
On cART at TB diagnosis (%)	26	25	34	8	<0.0001
On cART 12 months after TB diagnosis (%)	77	71	75	31	<0.0001

CONCLUSION: In conclusion, there were substantial differences in the clinical management of HIV-TB co-infected patients across Europe and Argentina, including less use of cART and more extensive use of second-line anti-TB drugs, presumably partly due to widespread TB drug resistance in populations from E. These factors may partly explain the 3–4 fold higher one-year mortality rate after a TB diagnosis in this region, and deserve immediate public health attention.

2008-11-10
1758-2652-11-S1-O39

This information is designed to support, not replace, the relationship that exists between you and your doctor.
1980, 2009. AEGiS.