Chinese Journal of Evidence -Based Pediatric ›› 2016, Vol. 11 ›› Issue (3): 187-194.

• Original Papers • Previous Articles     Next Articles

Diagnostic value of specimens via bronchoscopy for tuberculosis in children: a systematic review and meta-analysis

FU Wen-long1,2, ZHANG Juan-Juan1, DAI Ji-hong2   

  1. 1 Key Laboratory of Developmental Diseases in Childhood, Ministry of Education, 2 Center of Respiratory Disorders of Children′s Hospital, Chongqing Medical University, Chongqing 400014, China
  • Received:2016-03-18 Revised:2016-06-25 Online:2016-06-25 Published:2016-06-25
  • Contact: DAI Ji-hong

Abstract:

Objective To systematically evaluate the diagnostic value of specimens via bronchoscopy for tuberculosis(TB) detection in children.Methods The PubMed,EMBASE, Ovid, Cochrane Library, CBM, CNKI and Wanfang database were systematically searched up to February 2016, and relevant references were manual screened. Two reviewers independently screened literatures according to the inclusion and exclusion criteria and extracted data, and assessed the risk of bias of included studies by QUADAS-2 tool. The sensitivity and specificity of test methods and specimens (bronchoalveolar lavage, bronchial brushing and tissue biopsy) via bronchoscopy were evaluated by gold standard(clinical and laboratory diagnosis). Meta-analysis was performed using Meta-Disc 1.4 software, and summary receiver operating characteristic curve (SROC) was drawn and area under the curve (AUC) and Q index were calculated. Deek′s funnel plots were drawn using Stata 12.0 software.Results A total of 14 studies were included involving 2 781 specimens (1 082 patients). ①Fourteen literatures reported bronchoscopy for the diagnosis of TB. The pooled sensitivity and specificity of bronchoscopy for tuberculosis detection were 0.38(95%CI:0.35-0.41)and 1.00(95%CI:0.99-1.00), respectively. Then subgroup analysis was made by the types of specimens and test methods. The pooled sensitivity and specificity of tissue biopsy were 0.51(95%CI:0.40-0.61)and 1.00(95%CI:0.81-1.00) , respectively; The pooled sensitivity and specificity of culture, acid fast bacilli on smear and PCR for bronchoalveolar lavage were 0.36(95%CI:0.32-0.40)vs 1.00(95%CI:0.99-1.00), 0.21(95%CI:0.15-0.28)vs 1.00(95%:0.98-1.00)and 0.58(95%CI:0.46-0.69)vs 1.00(95%CI:0.84-1.00), respectively. ②Ten literatures reported gastric lavage for the diagnosis of tuberculosis, which exhibited heterogeneity(P<0.001,I2=93.3%). The pooled sensitivity and specificity of gastric lavage for tuberculosis detection were 0.45(95%CI:0.40-0.49)and 1.00(95%CI:0.99-1.00), respectively. The pooled sensitivity and specificity of smear for gastric lavage were 0.52(95%CI:0.47-0.57)and 1.00(95%CI:0.98-1.00), respectively. ③The SROC AUC of bronchoalveolar lavage and gastric lavage were 0.883 9 and 0.873 7, respectively; ④P values of the Deek′s test of culture and smear for bronchoalveolar lavage were 0.005 and 0.014. P value of the Deek′s test of gastric lavage was 0.147.Conclusion The value of bronchoscopy for the diagnosis of tuberculosis was definitive in children. Bronchoalveolar lavage for bronchoscopy was the most common specimen, but in terms of culture and smear, it wasn′t better than gastric lavage for tuberculosis detection.

CLC Number: