Background: Diagnosis of TB is still a challenge for paediatricians, physicians, radiologists, microbiologists, and serologists1. Clinical scoring systems are at present the most widely used method for arriving at the diagnosis of TB2. According to official estimates, the rate of open bacillary cases among adult population (18 years and above) was 17% and among children 5-9 years of age 13% were infected with TB. It is thought to be the 4th major cause of all deaths in Pakistan3. Objectives: To check the specificity and sensitivity of available scoring system of PPA and NTP clinical guidelines to diagnose TB in children. Study design: Descriptive Cross sectional study. Setting: Department of Peadiatrics,liaquat University Hospital Hyderabad/ Jamshoro. Duration with dates: Six months (17th January to 16th July 2009) Subjects and methods: In this study 100 patients of age group in between 6 months to 12 years, clinically suspected as TB on the basis of history of chronic cough, unexplained fever, weight loss, history of close contact with TB patient in family, has been evaluated for specificity and sensitivity of PPA and NTP guidelines for diagnosis of TB in children. As there is no specific tool to confirm the diagnosis so we confirm our diagnosis by assessing the response of anti-tuberculous drugs in all suspected cases of TB on the basis of scoring system. Data of each patient was collected on preformed Performa and analyzed on SPSS version 10. Frequency and percentages of different variables were computed. Mean and standard deviation was computed for quantitative variables like age of child. Chi-square test was applied to compare different proportion at level of significance of p <0.05. Results: The sensitivity and specificity of PPA was 91.3% and 50%, while in NTP guidelines sensitivity and specificity was found to be 90.1% and 44.44% respectively. Conclusion: PPA and NTP guidelines are more sensitive, but less specific parameter for the diagnosis of TB in children.