Letter to the Editor

To the Editor,

Haemophilus influenzae is a strictly human parasite responsible for a wide variety of respiratory and invasive infections. Six different serotypes (a through f) have been identified, as well as nontypeable strains that lack a capsule and, hence, the serotyping antigen. Previously, most invasive H influenzae diseases were due to serotype b strains; since the introduction of the conjugate H influenzae type b vaccines, the epidemiology of invasive H influenzae disease has changed tremendously, with most diseases now caused by either non-type b strains or nontypeable strains (1). Haemophilus parainfluenzae, on the other hand, is not usually regarded as a common pathogen, but has been known to cause infections such as pneumonia, peritonitis, biliary tract infection, endocarditis, urinary tract infection and exacerbation of chronic obstructive pulmonary disease (2).

Biotyping has been used for more than 30 years to characterize isolates of H influenzae and H parainfluenzae. Biotypes are assigned based on the biochemical reactions for indole, urease and ornithine decarboxylase production. Eight biotypes exist for each species, designated biotype I through biotype VIII (3–6). Despite having limited discriminatory power, when used in conjunction with other methods including newer genetic tools, biotyping may be helpful in the clinical laboratory to identify Haemophilus aegyptius (biotype III), which causes bacterial conjunctivitis (7), and Haemophilus quentini (biotype IV), which causes urogenital and neonatal infections (8).

Guidelines for biotype assignments can be found in several peer-reviewed publications and text books including Topley and Wilson’s Microbiology and Microbial Infections and the Manual of Clinical Microbiology (3–6,9–13). However, in the 10th edition of the Manual of Clinical Microbiology (13), the table showing biochemical reactions of H influenzae and H parainfluenzae biotypes contains a significant error that may lead to misidentification of biotypes. The error appears to involve the urease and ornithine decarboxylase reactions. Either the table subheadings or the results of the urease and ornithine decarboxylase reactions in the table content have been switched, leading to misidentification of H influenzae biotypes II, III, V and VI; and H parainfluenzae biotypes I, III, VI and VII. The remaining biotypes are not affected by this error because the results for urease and ornithine decarboxylase are identical, either both positive or both negative. As a result of this error, H influenzae biotype II would be misidentified as biotype V, biotype III would be misidentified as biotype VI, and biotypes V and VI would be misidentified as biotypes II and III, respectively (Table 1).

TABLE 1.

Comparison of Haemophilus biotype assignment tables

	Correct reactions*			Incorrect reactions†
Biotype	Indole	Urease	Ornithine decarboxylase	Indole	Urease	Ornithine decarboxylase
	Haemophilus influenzae
I	+	+	+	+	+	+
II	+	+	−	+	−	+
III	−	+	−	−	−	+
IV	−	+	+	−	+	+
V	+	−	+	+	+	−
VI	−	−	+	−	+	−
VII	+	−	−	+	−	−
VIII	−	−	−	−	−	−
Haemophilus parainfluenzae
I	−	−	+	−	+	−
II	−	+	+	−	+	+
III	−	+	−	−	−	+
IV	+	+	+	+	+	+
V	−	−	−	−	−	−
VI	+	−	+	+	+	−
VII	+	+	−	+	−	+
VIII	+	−	−	+	−	−

^*Kilian M. The genus Haemophilus. A taxonomic study of the genus Haemophilus, with the proposal of a new species. J Gen Microbiol 1976;93:9–62 (reference 3);

^†Ledeboer NA, Doern GV. Haemophilus. In: Manual of Clinical Microbiology, 10th edn. Washington, DC: ASM Press, 2011:588–602 (reference 13). − Negative; + Positive

Because biotypes I, II and III are the most common biotypes of H influenzae, (4–6,14,15), using the newer identification table in the 10th edition of the Manual of Clinical Microbiology could result in incorrect biotyping results for the majority of H influenzae strains. For example, at the National Microbiology Laboratory (Winnipeg, Manitoba), 19.4%, 56.2% and 16.7% of the H influenzae isolates analyzed in the past few years (n=1264) belonged to biotypes I, II and III, respectively (National Microbiology Laboratory, unpublished data). Therefore, using the new identification table would lead to the majority (72.9%) of these H influenzae isolates being misidentified. As of April 1, 2013, there have been no corrections issued by either the Manual of Clinical Microbiology or American Society for Microbiology. In summary, we wish to alert and remind our clinical microbiology colleagues of this error and the usefulness of this phenotypic method to raise the suspicion of the potential identification of unusual or uncommon organisms such as H aegyptius and H quentini.

Michelle Shuel BSc
Raymond SW Tsang PhD
Vaccine Preventable Bacterial Diseases
National Microbiology Laboratory
Public Health Agency of Canada
Winnipeg, Manitoba