Risk factors for drug resistant tuberculosis in southeast Turkey

Abstract

SUMMARY

We undertook a cross-sectional survey of 116 patients at Dicle Hospital, Turkey, who had with bacteriologically confirmed tuberculosis (TB). Demographic and clinical features, including age, gender, pulmonary TB history, associated diabetes mellitus, previous TB treatment, residential area and education, were collected from charts. Eighty-four of the strains were found to be susceptible to all drugs. The resistance to one or more drug(s) was found in 32 strains. Multi-drug resistant (MDR) TB was found in 13 strains (11.3% of the total and 40.7% of the drug resistant strains). The resistance to isoniazid was the most frequently seen (25 strains, 21.5%). In the multivariable analysis, only previous TB treatment (P = 0.000) remained a significant predictor for drug resistance; in MDR, previous TB treatments (P = 0.002) remained significant in the final model. The patient's educational status was found to be negatively correlated with the risk of MRD-TB (P = 0.035). Previous TB treatment and low educational status were found to important risk factors for the development of MDR-TB.

Introduction

Three countries – China, India and the Russian Federation – accounted for 261,362 multi-drug resistant tuberculosis (MDR-TB) cases or 62% of the estimated global burden.¹ Turkey is one of the moderate TB incidence countries where TB remains a public health threat and an economic burden. The incidence and mortality of TB decreased in Turkey at the end of the 20th century and the start of the new millennium with a 5.3/100.000 mortality rate and a 28/100.00 TB incidence rate in 2004.² The aim of this study was to assess risk factors for drug resistant TB (DR-TB) and MDR-TB at a university hospital in South-East Turkey.

Material and methods

The retrospective survey studied 116 patients with bacteriologically confirmed TB cases – all of the isolated M. tuberculosis in the Dicle University Hospital laboratory. All the records were reviewed for culture-proven TB cases. All drug susceptibility test results of the TB cases seen between December 2001 and April 2005 were included in the study. The testing of the susceptibility of the TB isolates to the four first-line antimicrobial drugs – isoniazid (H), rifampicin (R), ethambutol (E) and streptomycin (S) – was performed by the indirect proportion method using the Lowenstein–Jensen medium. The following concentrations were used: S, 2.0 µg/mL; H, 0.1 µg/mL; R, 2.0 µg/mL and E, 2.5 µg/mL.

DR was defined as resistance to one or more first-line drugs; multidrug resistance was defined as resistance to both H and R, with or without resistance to other drugs. A patient who had not received prior anti-TB treatment for more than one month was classified as a new case. Treatment failure was defined as failing an anti-TB treatment, i.e. having begun treatment for smear positive pulmonary TB and remaining smear positive at five months or later during the course of treatment.

Results

A total of 116 M. tuberculosis strains were isolated (Table 1). H was the most common of the drugs to which TB was resistant (Table 1). The sputum smear was positive in 55 (47.4%) of the cases (40 were sensitive to the drug and 15 were DR, of which six were in the MDR group). Ten (8.6%) had diabetes mellitus (DM) as a TB-associated disease. There was a family history in 25 (21.6%) patients. Of the 32 resistant cases, 17 were newly treated cases and 15 had previously received treatment for TB. Eight of the 116 patients had an inadequate treatment history, 17 had interrupted TB treatment (eight from the susceptible group and nine from the resistant group).

Table 1

The demographic features of patients

Variables	n (%)
Age ± standard deviation (years)	36.2 ± 21.1
Gender
Male	76 (65.5)
Female	40 (34.5)
Residential area
City centre	65 (56.0)
Small town	40 (34.5)
Rural area	11 (9.5)
^aEducational status
Illiterate	61 (52.5)
Literate	6 (5.2)
Primary school	35 (30.2)
High school or upper	14 (12.1)
Type of disease
Pulmonary TB	83 (71.5)
Tuberculous meningitis	14 (12.1)
Miliary TB	5 (4.3)
Others	14 (12.1)
History of TB in the family
Yes	25 (21.5)
No	91 (78.5)
Individual TB history
Yes	25 (21.5)
No	91 (78.5)
Drug resistance
Sensitive to all four drugs	84 (72.6)
Mono DR
H	5 (4.3)
R	0 (0.0)
E	0 (0.0)
S	5 (4.3)
Total monoresistance	10 (8.6)
Multidrug resistance
H + R	5 (4.3)
H + R + E	5 (4.3)
H + R + S	1 (0.8)
H + R + E + S	2 (1.9)
Total multidrug resistance	13 (11.3)
Other resistance patterns
H + E	1 (0.8)
H + S	5 (4.3)
H + E + S	1 (0.8)
E + S	1 (0.8)
R + E	1 (0.8)
Total other resistance patterns	9 (7.5)
Total	116 (100.0)

H, isoniazid; R, rifampicin; E, ethambutol; S, streptomycin

^aIlliterate and literate without school attending were accepted as uneducated

Previous TB treatment and an insufficient treatment history were associated with an increased risk of DR-TB (Table 2). In multivariable analysis, only previous TB treatment remained a significant predictor for drug resistance (adjusted odds ratio 10.33, 95% confidence interval 3.53–30.50, P = 0.000).

Table 2

Univariate analysis of potential predictors of drug resistant tuberculosis (DR-TB) and multi-drug resistant tuberculosis (MDR-TB)

Variables	Proportion of DR-TB	Proportion of sensitiveisolates	OR (95% CI)	P value
Age	36.3 ± 21.8	35.9 ± 19.5	–	0.203
Gender (male)	22/32 (68.8)	54/84 (64.3)	1.22 (0.51–2.92)	0.651
Education	45/84 (53.6)	39/84 (46.4)	0.52 (0.22–1.24)	0.139
Previously treated	15/32 (46.9)	10/83 (12.0)	6.44 (2.47–16.80)	0.000
Residential area (urban)	16/32 (50.0)	49/84 (57.1)	0.71 (0.32–1.62)	0.419
Diabetes mellitus	4/32 (12.5)	5/80 (6.3)	2.14 (0.14–1.74)	0.272
Unsuitable treatment	7/28 (25.0)	3/74 (4.1)	0.19 (0.04–0.87)	0.021
Adherence to treatment	4/23 (17.4)	7/73 (9.6)	1.99 (0.53–7.51)	0.306
Univariate analysis of potential predictors of MDR-TB cases
	Proportion of MDR	Proportion of non-MDR
Age	36.4 ± 21.4	34.6 ± 19.2	–	0.536
Gender (male)	7/13 (53.8)	69/103 (67.0)	0.58 (0.18–1.84)	0.347
Education	3/13 (23.1)	46/103 (44.7)	0.37 (0.97–1.43)	0.138
Previously treated	8/13 (61.5)	17/102 (10.8)	8.00 (2.33–27.45)	0.000
Residential area (urban)	9/13 (69.2)	56/103 (54.4)	1.89 (0.55–6.53)	0.309
Diabetes mellitus	3/13 (23.1)	6/99 (6.1)	4.65 (1.01–21.51)	0.341
Unsuitable treatment	1/11 (9.1)	7/91 (7.7)	1.20 (0.13–10.78)	0.871
Adherence to treatment	2/10 (20.0)	9/86 (10.5)	2.14 (0.39–11.67)	0.370

OR, odds ratio; CI, confidence interval

For MDR, also showed that only previous TB treatment regime was associated with an increased risk of DR (Table 2). Previous TB treatment (P = 0.002) remained a significant factor in the final model. Education was found to be an associated factor in the prevention of MDR (P = 0.035) (Table 2).

Discussion

A previous study reported that drug resistance was high (40.2%) in southeast Turkey.³ The report found that 60.8 % of the isolated strains were responsive to the drugs but 39.2% were resistant to at least one drug. In another study, total resistance to I, R and S was determined as 10.5%, 6.9% and 7.0%, respectively.⁴ Two studies from Turkey reported MDR-TB as 5.8% in one and 7.3% in the other.^4,5

The risk of MDR-TB was reported to be 10.23 times higher in previously treated patients than in those who had never been treated. In addition, MDR-TB cases more likely to be found in foreign born men aged less than 65years old who were HIV-positive.⁶ In our study, being uneducated was not a risk factor for DR but it was a significant risk factor for MDR. This variable was significant and independent of adherence to treatment and inadequate treatment. In a Korean study, the patient's level of education was reported as an associated factor for DR because of its influence on adherence to treatment.⁷ Among MDR-TB patients in Abidjan, 62% had an educational level of lower or equal to primary school standards.⁸

A Turkish study reported that symptoms, radiographic findings, associated diseases and sputum smear positivity had no effect upon the development of DR.⁹ Some studies showed that co-morbidity could be an important risk factor for DR. In one study, treatment results of MDR-TB were not satisfactory and diabetes was a risk factor for unfavourable results.¹⁰ Although MDR-TB patients are an epidemiological threat to the community improving their management significantly can be difficult.

References

Zignol

, Hosseini

, Wright

, Global incidence of multidrug-resistant tuberculosis. J Infect Dis 2006;194:479–85

WHO. Global Tuberculosis Control: Surveillance, Planning, Financing. WHO report 2006. Geneva: World Health Organization (WHO/HTM/TB/2006.362)

Balci

, Dikensoy

, Bayram

, Drug-resistant tuberculosis at the University Hospital in Gaziantep, South-eastern Turkey. J Int Med Res 2000;28:300–6

Bengisun

, Karnak

, Palabiyikoglu

, Saygun

. Mycobacterium tuberculosis drug resistance in Turkey, 1976–97. Scand J Infect Dis 2000;32:507–10

Kart

, Altın

, Tor

, Antituberculosis drug resistance patterns in two regions of Turkey: a retrospective analysis. Ann Clin Microbiol Antimicrob 2002;16:1–6

Faustini

, Hall

, Perucci

. Risk factors for multidrug resistant tuberculosis in Europe: a systematic review. Thorax 2006;61:158–63

Lee

, Chang

. Drug-resistant tuberculosis in a tertiary referral teaching hospital of Korea. Korean J Intern Med 2001;16:173–9

Kouassi

, Horo

, N'douba

, Epidemiological, clinical and biological profile of resistant or recurrent pulmonary tuberculosis in Abidjan. Bull Soc Pathol Exot 2004;97:336–7

Surucuoglu

, Ozkutuk

, Celik

, Drug-resistant pulmonary tuberculosis in western Turkey: prevalence, clinical characteristics and treatment outcome. Ann Saudi Med 2005;25:313–8

10.

Yoshiyama

, Ogata

, Wada

. Treatment results of multi drug resistant tuberculosis, a hospital based study. Kekkaku 2005;80:687–93