Portal de Periódicos da CAPES

Extensively drug-resistant tuberculosis (XDR-TB) among health care workers in South Africa

2010; Wiley; Volume: 15; Issue: 10 Linguagem: Inglês

10.1111/j.1365-3156.2010.02590.x

1365-3156

Julie Jarand, Karen Shean, Max R. O’Donnell, Marian Loveday, Charlotte Kvasnovsky, Martie van der Walt, Shahieda Adams, Paul A. Willcox, Justin O’Grady, Alimuddin Zumla, Keertan Dheda,

Pneumonia and Respiratory Infections

Objective To determine the clinical profile and outcomes of health care workers (HCWs) with extensively drug resistant tuberculosis (XDR-TB) in the Eastern and Western Cape Provinces of South Africa. Method Retrospective case record review of 334 patients with XDR-TB reported during the period 1996–2008 from Western and Eastern Cape Province, Cape Town, South Africa. Case records of HCWs with XDR-TB were analysed for clinical and microbiological features, and treatment outcomes. Results From 334 case records of patients with XDR-TB, 10 HCWs were identified. Eight of ten were HIV-uninfected, and four of 10 had died of XDR-TB despite treatment. All 10 HCWs had received an average of 2.4 courses of TB treatment before being diagnosed as XDR-TB. Conclusions In the Eastern and Western Cape provinces of South Africa XDR-TB affects HCWs, is diagnosed rather late, does not appear to be related to HIV status and carries a high mortality. There is an urgent need for the South African government to implement WHO infection control recommendations and make available rapid drug susceptibility testing for HCWs with suspected multidrug-resistant (MDR)/XDR-TB. Further studies to establish the actual risk and sources of infection (nosocomial or community) are required. Tuberculose ultra-résistante (TB-UR) chez les agents de la santéà Cape Town, Afrique du Sud Objectif: Déterminer la prévalence de la TB-UR chez les agents des soins de santé en Afrique du Sud. Méthode: Revue rétrospective des dossiers de cas portant sur 334 cas de TB-UR signalés au cours de la période de 1996 à 2008 dans l'ouest et l'est de la province du Cap, Cape Town, Afrique du Sud. Les dossiers des cas de TB-UR chez les agents de la santé ont été analysés pour les caractéristiques microbiologiques et cliniques et les résultats du traitement. Résultats: Sur 334 dossiers de cas de patients à TB-UR, 10 agents de la santé ont été identifiés. 8 des 10 n'étaient pas infectés par le VIH et 4 des 10 sont morts de TB-UR malgré le traitement. Tous les 10 agents de la santé avaient reçu une moyenne de 2,4 régimes de traitement antituberculeux avant d'être diagnostiqués comme TB-UR. Conclusions: A Cape Town, la TB-UR affecte les agents des soins de santé, est diagnostiquée assez tard, ne semble pas être liée au statut VIH et comporte une forte mortalité. Il est urgemment nécessaire que le gouvernement d'Afrique du Sud implémente les recommandations de l'OMS pour la prévention des infections et rende disponible des tests rapides de sensibilité pour les patients et les agents de la santé ayant une suspicion de TB-MDR/UR. D'autres études pour établir le risque réel et les sources d'infection (nosocomiale ou communautaire) sont nécessaires. Tuberculosis ampliamente resistente a medicamentos (XDR-TB) entre trabajadores sanitarios en Ciudad del Cabo, Sudáfrica Objetivo: Determinar la prevalencia de XDR-TB entre trabajadores sanitarios en Sudáfrica. Método: Revisión retrospectiva de casos en 334 casos de XDR-TB reportados durante el período entre 1996 y 2008 en la provincia Occidental y Oriental del Cabo, Ciudad del Cabo, Sudáfrica. Se analizaron las historias clínicas de los casos de XDR-TB de trabajadores sanitarios en busca de características clínicas y microbiológicas, así como los resultados del tratamiento. Resultados: De 334 historias clínicas de pacientes con XDR-TB, se identificaron 10 trabajadores sanitarios. 8 de 10 eran VIH-negativos y 4 de 10 habían muerto por XDR-TB a pesar del tratamiento. Los 10 trabajadores sanitarios recibieron en promedio unos 2.4 cursos acerca del tratamiento de la TB antes de ser diagnosticados como XDR-TB. Conclusiones: En Ciudad del Cabo, la XDR-TB afecta a trabajadores sanitarios, se diagnostica relativamente tarde, no parece estar relacionada con el seroestatus para VIH, y tiene una alta mortalidad. Hay una necesidad urgente de que el gobierno sudafricano implemente las recomendaciones de control de la infección de la OMS, y que haga disponible el testaje rápido para la susceptibilidad a medicamentos para pacientes y trabajadores sanitarios con sospecha de TB multirresistente o XDR-TB. Se requieren más estudios para establecer el riesgo actual y las fuentes de infección (nosocomial y comunitario). Multidrug (MDR)- and extensively drug (XDR)-resistant tuberculosis (TB) carry a high mortality rate and are a growing problem that threatens to destabilize tuberculosis control globally (WHO Report. World Health Organization, 2010; WHO, 2010; Centers for Disease Control and Prevention, 2006; Sotgiu et al. 2009; Gandhi et al. 2006). South Africa has a very high annual TB incidence of 948 per 100 000 population/year with approximately 8000–9000 patients per annum with multidrug-resistant TB (MDR-TB) passively detected per annum. The most recent report of 5% of the patients with MDR-TB being extensively drug-resistant TB (XDR-TB) in South Africa is likely an underestimate (WHO Report. World Health Organization, 2010; WHO, 2010). TB in all its forms is having a devastating effect on the world's population and poses a great risk to the health and work outputs of health care workers (HCWs) (Menzies et al. 2007; Naidoo & Jinabhai 2006; Jo et al. 2008; Pillay & Malhati 2008; Galgalo et al. 2008; Joshi et al. 2006). Health care workers in South Africa are at the forefront of the battle against the TB and HIV/AIDS epidemics (Naidoo & Jinabhai 2006; Joshi et al. 2006; Shisana et al. 2004). The shortage of HCWs has reached alarming levels in sub-Saharan Africa, which is home to 11% of the world's population and 24% of the global burden of disease, but only 3% of the world's HCWs (Wilkinson & Gilks 1998). Multiple factors contribute to this problem, such as high workload, poor remuneration, limited training opportunities and lack of resources (Pillay & Malhati 2008). Other significant contributing factors are fatalities and incapacitating disability related to HIV/AIDS and other infectious diseases of poverty. In 2002, an estimated 16% of HCWs in South Africa were infected with HIV (Shisana et al. 2004). South Africa has one of the highest rates of TB and HIV/AIDS in the world, and the occurrence of TB in HCWs has been reported to be substantial (Naidoo & Jinabhai 2006; Shisana et al. 2004; Wilkinson & Gilks 1998). However, the occurrence and importance of XDR-TB in HCWs in South Africa are not yet documented. Health care workers with active tuberculosis (TB) need to be identified and treated. HCWs with active TB risk infecting their patients, their fellow staff, families and communities with Mycobacterium tuberculosis (Mtb). Conversely, HCWs are also at risk of acquiring TB from their patients, fellow staff and family and community members. Over the past 15 years, the risk of TB in HCWs in southern Africa has been increased because of the resurgence of TB caused by the HIV epidemic and the emergence of MDR-/XDR-TB, which has added significant strain to already overwhelmed TB control programmes (Gandhi et al. 2006; Naidoo & Jinabhai 2006; Shisana et al. 2004). The prevalence of extensively drug-resistant tuberculosis (XDR-TB) is now being increasingly reported from South Africa (Gandhi et al. 2006; Dheda et al. 2010) and poses a major threat to HCWs and other patients in the wards and outpatient clinics, and those in congregate settings. XDR-TB is difficult to treat and has poor treatment outcomes, and together with HIV co-infection, may exacerbate the shortage and poor performance of HCWs in South Africa. The frequency, clinical profile and outcomes of HCWs with XDR-TB in South Africa remain unknown (Sotgiu et al. 2009). To ascertain whether XDR-TB is an emerging problem in HCWs in South Africa, we undertook a retrospective case record review of 334 patients with XDR-TB reported during the period 1996–2008 from the Eastern and Western Cape Provinces of South Africa. Using a validated data capture tool, we performed a retrospective study of case records of 334 patients with XDR-TB, diagnosed between January 1996 and February 2008, from the Western Cape and Eastern Cape provinces of South Africa. For the HCWs identified with XDR-TB, patient data, including demographics, microbiologic and treatment records, were collected. The diagnosis of XDR-TB was based on sputum culture positivity for Mtb bacilli, and was defined, based on drug susceptibility testing (DST), as resistant to isoniazid, rifampicin, any one fluoroquinolone and at least one-second-line injectable anti-TB drug. Case definitions for diagnosis and conversion are outlined in Table 1. Ten HCWs with XDR-TB were identified. Table 1 outlines patient characteristics, including gender, HIV status, occupation, place of work, number of previous TB episodes, chest X-ray abnormalities, number of resistant drugs, treatment accorded and treatment outcomes. Patients were predominantly female (9 of 10) with a median age of 40 years (range 26–50). The majority of patients (6 of 10) worked as nursing staff and one each in the following professions: doctor, radiographer, laundry staff and ward cleaner. Of the 10 patients, eight were HIV-uninfected. One HIV-positive patient had been treated with anti-retrovirals (ARVs) for 4 years prior to diagnosis of XDR-TB, and the other HIV-uninfected patient was started on ARVs at the time of XDR-TB diagnosis. All 10 patients had been treated for TB previously with an average of 2.4 previous TB episodes. The median duration of follow-up was 10 months (range 1–22 months) from the time of XDR-TB diagnosis. Table 2 shows the individual patient characteristics (age, gender, year of diagnosis, occupation, HIV status, drug-resistant patterns and treatment accorded and treatment outcome) of the 10 HCWs with XDR-TB that were studied. Three patients underwent surgical resection because of failure to improve clinically, and they had unilateral disease. A high mortality was observed (4 of 10 patients died). Para-amino-salicylic acid (PAS) and capreomycin became available again in South African government-run hospitals in late 2006. Seven patients were treated with regimens containing these drugs, two patients were diagnosed and died before these drugs were available and one patient responded to a modified MDR-TB regimen which did not include capreomycin or PAS. Four patients died and six were still being maintained on treatment for XDR-TB at the end of the study. This case series study adequately answers our primary study aim, confirming that XDR-TB in HCWs is an important clinical problem, and should be of major concern to TB programme managers in South Africa. It affects a spectrum of HCWs, particularly nurses in the Western and Eastern Cape Provinces of South Africa. Whilst in MDR-TB sputum culture conversion is usually achieved in most patients within 3–4 months (WHO, 2010) of initiation of MDR-TB therapy, six of the 10 HCWs with XDR-TB in this study did not achieve culture conversion within this time frame. Moreover, two of the four converters reverted back to culture positivity prior to the end of follow-up, and the overall mortality was high (40%). Thus, treatment-related outcomes seen in this limited case series of 10 patients were poor. This is concordant with other observations on treatment success (cured and completed) ranging from 34% to 67% and mortality of 7–36% in non-HCW populations (Sotgiu et al. 2009). Health care workers acquiring XDR-TB adds to the growing burden of drug-resistant disease. It also concurrently exacerbates the shortage of HCWs as they can no longer work while undergoing prolonged inpatient treatment and during convalescence (Pillay & Malhati 2008; Wilkinson & Gilks 1998; Dheda et al. 2010; World Health Organization, 2006). This emphasises the need for intensified HCW screening, the widespread use of newer tools now available for the rapid diagnosis of MDR-/XDR-TB and instituting appropriate effective treatment. Importantly, the delay in diagnosis of XDR-TB in these HCWs indicates that undiagnosed and untreated, they are a risk to inpatients they manage, other fellow staff, and their families and communities. Because of the limitations of our study design and the data obtained, it is unclear where and from whom the HCWs acquired the infection: in hospital as a nosocomial infection from inpatients with XDR-TB they cared for or from the community where XDR-TB is becoming more prevalent. Nosocomial transmission of drug-resistant Mtb strains has been well documented and therefore HCWs are clearly at risk (Gandhi et al. 2006; Menzies et al. 2007). Given that HCWs in our study were previously treated for TB, (including on at least two previous TB episodes in eight patients), community acquired, rather than nosocomial, XDR-TB may be a possibility. However, given the high prevalence of previous TB in patients with drug resistance in the Western Cape (Dheda et al. 2010) and studies demonstrating re-infection as an important modality of transmission (Verver et al. 2005), it is imperative that the transmission dynamics of these infections be ascertained. However, without identification of the Mtb strains and their molecular typing data from the HCWs studied, and from the index case, and from prior TB episodes, it was not possible to determine the transmission dynamics in the cases that we describe. The transmission patterns of drug-resistant Mtb strains and whether XDR-TB may harbor different Mtb isolates and strains are unknown. Recent observations that Mtb strains have a rapid rate of molecular evolution at the hypervariable and immunogenic Mtb PPE38 gene region, which may lead to evasion by the immune system, may be important. It may also be that Mtb strains may be more diverse than previously thought (McEvoy et al. 2009). Prospective studies are now required to address these important issues for proper interventions to be developed. Although XDR-TB in South Africa was initially thought to be almost exclusively associated with HIV coinfection (Gandhi et al. 2006), this study shows that HIV-uninfected HCWs are also at risk of acquiring XDR-TB. Therefore, MDR-/XDR-TB should be considered in any HCW suspected of TB, irrespective of HIV status, and all HCW should be screened with the newer rapid first- and second-line drug susceptibility tests if they are found to be smear or culture positive (Pai et al. 2006). All HCWs in South Africa should also be encouraged to be tested for HIV, and if HIV-infected, they should receive appropriate treatment for TB and HIV/AIDS. Upon recovery, they should also be relocated to work in areas with lowest risk of TB and in health settings where appropriate WHO-recommended infection control measures are being implemented. The obvious limitations of our study, given its retrospective design and ascertainment bias, are the inability to calculate prevalence estimates, or to precisely delineate whether the XDR-TB in HCWs was acquired nosocomially or from the community. The physical, social and economic consequences of HCWs acquiring XDR-TB can be very severe and affect their daily lives. Thus, it becomes important for the South African government to take immediate measures to provide the resources required to enable all hospitals and clinics providing TB care to follow and to implement the recommended WHO infection control procedures (WHO Report, 2009). This will go a long way in protecting HCWs and other inpatients from acquiring MDR-/XDR-TB from those patients and HCWs who have active disease. Furthermore, it is important that all health facility laboratories be equipped with the newer rapid diagnostic testing platforms now available, so that all patients and HCWs suspected of having TB need can be identified early, and appropriately isolated, to minimize the risk of transmission within hospitals and in the community. In the Western and Eastern Cape provinces of South Africa XDR-TB is diagnosed rather late, does not appear to be related to HIV status and carries a high mortality. This underscores the need for the government to implement current WHO policy on TB infection control in all health care facilities in South Africa. There is also the need for making available rapid molecular biological DST for all HCWs and inpatients to rapidly detect, isolate, and treat all patients with MDR-/XDR-TB. Now that we have shown that HCWs succumb to XDR-TB, further studies to establish the source and mode of infection in HCWs, and the risk to their patients and families, are urgently required. Keertan Dheda was supported by the South African National Research Foundation, the South African Medical Research Council, an EU FP7 Grant and the EDCTP. Alimuddin Zumla receives support from the EU, EuropeAID, EDCTP, UK-MRC and the NIHR UCL-HNHSTrust CBRC. Charlotte Kvasnovsky and Martie Van der Walt were supported by the South African MRC and Center for Disease Control through a cooperative agreement to MRC. Charlotte Kvasnovsky was supported by a grant from Emory School of Medicine.