Detection of virus in water: sensitivity of the tentative standard method for drinking water

The sensitivity of several microporous virus-adsorbent media for reliably detecting low levels of poliovirus from 380 and 1,900 liters of drinking water by use of the tentative standard method was investigated. The virus-adsorbent media tested were (i) nitrocellulose membrane filters, (ii) epoxy-fiber glass-asbestos filters, (iii) yarn-wound fiber glass depth filters, and (iv) epoxy-fiber glass filter tubes. Virus was adsorbed to the filter media at pH 3.5 and eluted with glycine buffer, pH 11.5. The results from 44 samples demonstrated that poliovirus was detected with a 95% reliability at mean virus input levels of 3 to 7 plaque-forming units/380 liters when 1,900 liters of water was sampled. At mean virus input levels of less than 1 to 2 plaque-forming units/380 liters, the detection reliability was 66% in 76 samples when 1,900 liters of water was sampled. No significant difference in virus detection sensitivity was observed among the various virus adsorbent media tested. Overall virus recovery efficiency ranged from 28 to 42%, with a grand average of 35%. Members of the coxsackievirus groups A and B, echovirus, and adenovirus were also detected when 380 and 1,900 liters of water were sampled. These experimental observations attest to the sensitivity of the tentative standard method for detecting low levels of virus in large volumes of drinking water.