Treatment Strategies for Group A Streptococcal Pharyngitis

In the recent, nicely done study by Needham et al. (2) concerning the impact of a rapid group A streptococcal antigen assay on physician usage of appropriate antibiotics, there are two areas of concern.

(i) In the abstract, the authors state that if the rapid antigen test alone had guided therapeutic choice, an appropriate course of antibiotics would have been prescribed for 95% of the patients during the initial visit. In their conclusion, last paragraph, the authors further state that neither the clinical assessment of the patients by their physicians nor culture added significantly to the improved outcome (i.e., appropriately using or withholding antibiotics) and that the preferred strategy was exclusive reliance on the rapid antigen test. While it is true that 95% of the total number of culture-positive and -negative patients would have been appropriately treated or not treated if the decision to treat had been based solely on antigen assay results, 15 (12.9%) of their 116 culture-positive patients (the target population for the diagnostic tests) would not have been treated if this strategy had been followed because their antigen test results were falsely negative. This is an unacceptably high rate of diagnostic failures and treatment errors.

In order to prevent the suppurative and nonsuppurative streptococcal sequelae mentioned in the work by Needham et al. (2), as well as to avoid unnecessary therapy, it still appears that a more effective treatment strategy would be to (a) treat all of those with positive antigen results and withhold therapy from as many as clinically possible of the rest pending culture results (since neither antigen assays nor culture can claim 100% sensitivity, the diagnosis of group A streptococcal pharyngitis should include both clinical and epidemiologic findings [3]), (b) perform cultures for all patients with negative streptococcal antigen results, and (c) treat the antigen-negative, culture-positive patients as currently recommended (3).

(ii) The authors reported that 11 of the 15 false-negative antigen test results came from patients whose cultures contained only rare to few colonies of group A streptococci (2). The implication may be that those patients with small numbers of streptococcal colonies from culture are only colonized, not infected. This interpretation would be in direct conflict with the conclusion of an excellent study by Gerber et al. (1) that found that the differentiation of patients with streptococcal infections from those who are only carriers of the organism could not be made on the basis of the degree of positivity of the culture alone.

TABLE 1.

Diagnostic tests performed and complications during each time period

Descriptor	No. of pharyngitis patients during:		P
	Period 1a	Period 2b
Diagnostic test(s)	15,399	14,637
Throat culture	10,098	2,087	<0.001
Antigen test	8	7,434	<0.001
Antigen test + culture	1	184
No test	5,292	4,932
Suppurative complications	56	43	NSc
Peritonsillar abscess or cellulitis	34	35
Retropharyngeal abscess	3	1
Pharyngitis and lymphadenitis	19	7
Nonsuppurative complications	0	1	NS
Acute rheumatic fever	0	0
Post-streptococcal infection glomerulonephritis	0	1

^a1 April 1994 to 31 March 1996. 

^b1 April 1996 to 31 March 1998. 

^cNS, not significant.