R T Hayden1, L D Frenkel. 1. Swedish American Hospital, University of Illinois College of Medicine at Rockford, 61104, USA.
Abstract
BACKGROUND: The bacterial latex agglutination assay is ordered predominantly on the pediatric population, for rapid screening for bacterial surface antigens in cerebrospinal fluid (CSF) or urine specimens. The high cost of this assay and questions raised in the literature regarding its accuracy led to a retrospective review of the use of this assay at a medium-sized midwest teaching hospital. The results of 6,370 bacterial latex agglutination tests performed between May, 1995, and November, 1996, and charts of patients being tested were reviewed. RESULTS: This study demonstrated a sensitivity and specificity of 28.6% and 86.7% for urine specimens and 70.0% and 99.4% for CSF specimens. A total of 11 pathogens were accurately detected (7 CSF and 4 urine). There were 13 false negatives and 59 false positives. None of the true positives had a discernible effect on either treatment or hospital course; however, several of the erroneous tests resulted in delayed or unnecessary treatment and workup of the involved patients. The annual billed cost of this test at this institution (fiscal years 1995 to 1997) averaged $167,000 per annum. This does not include indirect costs associated with increased length of hospital stay, overutilization of antibiotics and excess laboratory tests ordered as a result of false positives. CONCLUSIONS: Bacterial antigen latex agglutination testing is neither sufficiently sensitive nor specific to be used as a screening test. Accurate results have no demonstrable clinical impact, whereas numerous inaccurate results are often generated at great cost. The continued use of the latex agglutination assay should be seriously questioned in an era when cost containment and clinical efficiency are becoming increasingly important.
BACKGROUND: The bacterial latex agglutination assay is ordered predominantly on the pediatric population, for rapid screening for bacterial surface antigens in cerebrospinal fluid (CSF) or urine specimens. The high cost of this assay and questions raised in the literature regarding its accuracy led to a retrospective review of the use of this assay at a medium-sized midwest teaching hospital. The results of 6,370 bacterial latex agglutination tests performed between May, 1995, and November, 1996, and charts of patients being tested were reviewed. RESULTS: This study demonstrated a sensitivity and specificity of 28.6% and 86.7% for urine specimens and 70.0% and 99.4% for CSF specimens. A total of 11 pathogens were accurately detected (7 CSF and 4 urine). There were 13 false negatives and 59 false positives. None of the true positives had a discernible effect on either treatment or hospital course; however, several of the erroneous tests resulted in delayed or unnecessary treatment and workup of the involved patients. The annual billed cost of this test at this institution (fiscal years 1995 to 1997) averaged $167,000 per annum. This does not include indirect costs associated with increased length of hospital stay, overutilization of antibiotics and excess laboratory tests ordered as a result of false positives. CONCLUSIONS: Bacterial antigen latex agglutination testing is neither sufficiently sensitive nor specific to be used as a screening test. Accurate results have no demonstrable clinical impact, whereas numerous inaccurate results are often generated at great cost. The continued use of the latex agglutination assay should be seriously questioned in an era when cost containment and clinical efficiency are becoming increasingly important.