On November 6, 1994, a 4-year-old girl was admitted to the Emergency Room of St. Mary's Hospital in Rogers, Arkansas, with fever and generalized lymphadenopathy. Two other children in her home day care, including the patient's sibling, also had lymph node enlargement and fever. Her mother reported that the patient had not been immunized against any of the common childhood diseases but had an unremarkable medical history, with the exception of hand-foot-and-mouth disease at 3 years of age; she had never received antibiotics. Her developmental milestones were achieved on time. The family included the patient, her mother, father, and a 3-year-old sibling.

The patient's posterior pharynx was erythematous with swelling of the posterior palate and with a thick grayish white membranous exudate adhering to the posterior wall of the pharynx. The pseudomembrane had almost completely detached and was gone, leaving behind ulcerated tissue. A discrete adenopathy in the neck and palpable nodes in the axillary, epitrochlear, and inguinal regions were observed.

The initial white blood cell count was 7,000 per ml, with 50% polymorphonuclear cells and 17% lymphocytes. A rapid screening test for group A streptococci and a Paul-Bunnell test for Epstein- Barr virus had negative results. Routine throat and nasopharyngeal swabs were collected and submitted to the hospital's clinical microbiology laboratory, where they were injected into blood agar and blood chocolate agar (special media for diphtheria culture were unavailable). From the throat swab, there was light growth of Streptococcus pneumoniae and light growth of gram-positive rods; a mixture of normal flora was also present. Gram-positive rods were also isolated from the nasopharyngeal swab. Subsequently, when subcultured on cystine-tellurite agar, these gram-positive rods grew as black colonies. Subcultures were then sent to the state laboratory, which confirmed them as C. pseudodiphtheriticum. After ambulatory treatment with cefprozil and erythromycin, the patient recovered in 4 days.

Since 1932, only 83 cases of disease possibly caused by C. pseudodiphtheriticum in humans have been reported (Table). Until 1981, the only reported disease associated with this organism was endocarditis in the presence of anatomic abnormalities of the heart. Since then, this organism has been increasingly recognized as a pathogen of the lungs and bronchi, particularly in patients with underlying immunosuppressive conditions and preexisting pulmonary diseases and in patients undergoing endotracheal intubation.

Among reported cases, 19 were endocarditis (4-17); 61 involved infections of the lungs, trachea, or bronchi (18-32); one was a urinary tract infection (33); one involved a skin infection (34); and one was a vertebral discitis (35). All but three patients had underlying medical conditions, including functional or anatomic abnormalities of the heart (N=27), lung and tracheobronchial diseases (N=28), endotracheal intubations (N=3), and immunosuppressive conditions including prolonged steroid use (N=6) and AIDS (N=4).

Most patients showed good clinical response to various antimicrobial drugs: penicillin, ampicillin, cefazolin, vancomycin, gentamicin, tobramycin, norfloxacin, and others. Nineteen (23%) deaths were reported overall. Information on sex was available for 80 cases; 54 (68%) were in males. Only five (6%) cases occurred in children <18 years of age; no cases were reported among children <5 years of age.

Since the early 1990s, diphtheria has returned in epidemic proportions in the former Soviet Union and is likely to be imported into other countries. Two cases of exudative upper respiratory tract infections with a pseudomembrane, in which the main organism isolated was C. pseudodiphtheriticum, have been reported. The first case was in a 54-year-old man with necrotizing tracheitis in 1991 (32). Santos et al. (36) reported isolation of C. pseudodiphtheriticum from a 32-year-old male Uzbek national who had a severe sore throat and dysphagia of 2 days duration. His tonsils were enlarged bilaterally, and a grayish-white exudate extended from the tonsil to the posterior pharyngeal wall. The uvula and soft palate were erythematous and edematous, and tender cervical lymphadenopathy was present. The presumptive clinical diagnosis in this patient was respiratory diphtheria.

Although the presence of a grayish pseudomembrane in the upper respiratory tract associated with isolation of C. pseudodiphtheriticum in these two cases and in the case we described is not proof that the isolated corynebacterium caused the pharyngitis or tracheitis, these accumulating case reports suggest that C. pseudodiphtheriticum may have played a pathogenic role.

The isolate from our patient was nontoxigenic by PCR. The diphtheria toxin is considered to be the main virulence factor of C. diphtheriae that causes the formation of the pseudomembrane in faucial diphtheria (37). Several possibilities could explain the presence of a pseudomembrane in infections with nontoxigenic corynebacterium strains. First, toxigenic and nontoxigenic colonies of the same species can coexist in a patient; this phenomenon has been reported in diphtheria (38). Multiple colonies would have to be tested to confirm this hypothesis in this case. Second, nontoxigenic forms of C. diphtheriae can produce invasive disease (39,40). It is possible that the symptoms in our patient were caused by a nontoxigenic C. pseudodiphtheriticum and that the pseudomembrane was simply an inflammatory exudate. This has been reported for Arcanobacterium haemoliticum (formerly Corynebacterium haemoliticum), which has also been associated with production of a grayish pharyngeal pseudomembrane (41,42). In 1960, Barksdale et al. reported a pseudomembrane in two laboratory workers contaminated with known nontoxigenic forms of C. diphtheriae (43). Third, we cannot exclude the possibility that the symptoms were caused by other organisms, including S. pneumoniae, which was found in the throat culture of our patient (29). However, this is unlikely because the growth of S. pneumoniae in our patient's throat culture was light, and the organism was not found in the nasopharyngeal swab. Furthermore, S. pneumoniae has not been reported to produce a pseudomembrane (44,45).

Because several other Corynebacterium spp. have been identified as potentially pathogenic for humans, routine screening for C. diphtheriae and other members of the Corynebacterium spp. in clinical samples of patients with respiratory infections should be encouraged (46). Full identification of gram-positive rods isolated from the respiratory tract, especially when they appear on original plates as the predominant organisms or copredominant with another species, should be carried out at the local, state, or reference level. Simple screening tests, such as negative cysteinase and positive pyrazinamidase, in addition to inability to ferment glucose, maltose, and sucrose quickly differentiate C. pseudodiphtheriticum from C. diphtheriae. These tests should be a part of improved training of laboratory personnel. Clinical and laboratory experience gained through this process will not only provide information about death rates for these organisms, but would be invaluable in preventing unnecessary concern in the community and the extraordinary measures needed to control dissemination of diphtheria.