Fischl M A, Uttamchandani R B, Daikos G L, Poblete R B, Moreno J N, Reyes R R, Boota A M, Thompson L M, Cleary T J, Lai S
University of Miami School of Medicine, Department of Medicine, FL 33101.
Ann Intern Med. 1992 Aug 1;117(3):177-83. doi: 10.7326/0003-4819-117-3-177.
To evaluate a nosocomial outbreak of tuberculosis caused by multiple-drug-resistant bacilli among patients with tuberculosis and HIV infection.
A case-control study.
Patients with HIV infection and culture-proven tuberculosis.
Patient characteristics, date of diagnoses of HIV infection and disease, date of diagnosis of tuberculosis, Mycobacterium tuberculosis susceptibility results, and medical center contact.
Sixty-two patients who had tuberculosis caused by multiple-drug-resistant bacilli (cases) and 55 patients who had tuberculosis caused by susceptible or single-drug-resistant bacilli (controls) were identified. Controls were more likely to be black (odds ratio, 0.4; 95% CI, 0.2 to 0.9) or Haitian (odds ratio, 0.2; CI, 0.1 to 0.6) compared with cases, who were more likely to be homosexual men (odds ratio, 2.9; CI, 1.3 to 6.4). Forty-four cases (71%) had previous contact with an HIV clinic compared with 15 controls (27%) (P less than 0.0001). Cases were more likely to have had AIDS (odds ratio, 7.7; CI, 1.5 to 53.7), to have been hospitalized on an HIV ward (odds ratio, 8.3; CI, 2.3 to 29.7), to have been seen in an HIV clinic (odds ratio, 7.8; CI, 3.4 to 18.1), to have received intravenous therapy in an HIV clinic (odds ratio, 13.0; CI, 4.6 to 37.0), or to have received inhalation pentamidine in an HIV clinic before a diagnosis of tuberculosis was made. Multiple logistic regression analysis showed that a diagnosis of AIDS (odds ratio, 11.2; CI, 3.1 to 40.6) and HIV clinic visits (odds ratio, 13.0; CI, 2.7 to 63.7) before a diagnosis of tuberculosis were significantly associated with tuberculosis caused by multiple-drug-resistant bacilli. Using susceptibility patterns and appointment dates, we found that 22 cases had previous contact with a person who had tuberculosis caused by multiple-drug-resistant bacilli in the HIV clinic.
Nosocomial transmission of M. tuberculosis from other HIV-infected patients with tuberculosis caused by multiple-drug-resistant bacilli can occur. These findings have serious public health implications and demand strict adherence to acid-fast bacilli isolation precautions.
评估结核病合并人类免疫缺陷病毒（HIV）感染患者中由多重耐药杆菌引起的医院内结核病暴发情况。
病例对照研究。
HIV感染且经培养证实患有结核病的患者。
患者特征、HIV感染及疾病诊断日期、结核病诊断日期、结核分枝杆菌药敏结果以及医疗中心接触情况。
确定了62例由多重耐药杆菌引起结核病的患者（病例组）和55例由敏感或单耐药杆菌引起结核病的患者（对照组）。与病例组相比，对照组更可能为黑人（比值比，0.4；95%可信区间，0.2至0.9）或海地人（比值比，0.2；可信区间，0.1至0.6），而病例组更可能为男同性恋者（比值比，2.9；可信区间，1.3至6.4）。44例病例（71%）此前曾接触过HIV诊所，而对照组为15例（27%）（P<0.0001）。病例组更可能患有获得性免疫缺陷综合征（AIDS）（比值比，7.7；可信区间，1.5至53.7），曾在HIV病房住院（比值比，8.3；可信区间，2.3至29.7），曾在HIV诊所就诊（比值比，7.8；可信区间，3.4至18.1），曾在HIV诊所接受静脉治疗（比值比，13.0；可信区间，4.6至37.0），或在结核病诊断前在HIV诊所接受过吸入性喷他脒治疗。多因素逻辑回归分析显示，结核病诊断前的AIDS诊断（比值比，11.2；可信区间，3.1至40.6）和HIV诊所就诊（比值比，13.0；可信区间，2.7至63.7）与多重耐药杆菌引起的结核病显著相关。根据药敏模式和就诊日期，我们发现22例病例此前在HIV诊所曾接触过由多重耐药杆菌引起结核病的患者。
结核分枝杆菌可在医院内从其他感染HIV且患有由多重耐药杆菌引起结核病的患者传播。这些发现具有严重的公共卫生意义，需要严格遵守抗酸杆菌隔离预防措施。
