All patients were admitted to local hospitals and clinical samples for microbiological investigations were sent to the nearest laboratories. C. diphtheriae isolates were sent to National Institute of Public Health–National Institute of Hygiene for confirmation and toxigenicity testing, biotyping, pulsed-field gel electrophoresis (PFGE), multilocus sequence typing (MLST), and ribotyping. Case classification and microbiological methods used are presented in the Technical Appendix. Data collected for epidemiologic analysis included location; type of infection; year of presentation; and patient age, sex, concomitant diseases, socioeconomic status, and intravenous drug use (IVDU).

All isolates from local and invasive infections were identified as biotype gravis, except for the isolate from patient Loc-05, which was identified as biotype mitis. All 25 isolates were characterized by PFGE, and 20 isolates (18 from invasive and 2 from local infections) were characterized by MLST (3–5). All the isolates except the mitis isolate belonged to the same pulsotype revealed by PFGE. All the isolates characterized by MLST belonged to genotype sequence type 8. Eight of the isolates (5 from invasive and 3 from local infections) also were genotyped by using ribotyping. All 8 isolates showed indistinguishable ribotype patterns (3).

Figure

Figure. . . Number of nontoxigenic Corynebacterium diphtheriae infections, Poland, 2004–2012. Excluded are 5 cases for which no data were available.

All but 1 invasive infection were identified in male patients, whereas local infections affected male and female patients similarly. Age groups of patients most affected by invasive infections were 31–40 years, followed by 51–60 years; for local infections, persons 51–60 years of age were mostly affected (Figure). Patients’ ages ranged from 17 to 71 years. The cases occurred in various parts of Poland; no epidemiologic links were identified.

Epidemiologic data analysis revealed that predisposing factors of nontoxigenic C. diphtheriae invasive infections were related to conditions associated with low socioeconomic status, such as alcoholism, homelessness, and dental caries, as well as to hepatic cirrhosis. For 3 (17%) patients (Inv-08, Inv-09, Inv-13), no concomitant diseases or predisposing factors were identified. These were healthy men aged 17, 24, and 37 years of age, respectively. Predisposing factors for local infections were not analyzed. The sources of all infections described in the study were not identified. Despite IVDU being regarded as a risk factor for C. diphtheriae invasive infection, none of the patients were intravenous drug users.

Diphtheria is a rare disease in Europe. In 2006–2009 only 150 cases were reported in European Union and European Economic Area/European Free Trade Association countries. Most of the cases (114 cases) occurred in Latvia, where diphtheria is endemic. The other diphtheria cases were reported in France (10 cases), Germany (6), Sweden (2), United Kingdom (16), and Norway (4) (6). However, infections from nontoxigenic C. diphtheriae have been reported in several European countries, such as Germany (7), United Kingdom (8), France (5), Switzerland (9), and Italy (10), during the past few years. In Poland, persons most affected were 31–40 years and 51–60 years of age, whereas in other countries most patients were younger (up to 34 years of age). No C. diphtheriae infections among children were recorded in Poland, whereas in France, almost 20% of invasive infections were diagnosed in children. On the other hand, in Italy and the United Kingdom, 70% and 13% of isolates, respectively, originated from patients <15 years of age (5,8,10).

In Poland, all but 1 strain isolated from local and invasive infections belonged to biotype gravis, whereas biotype mitis dominated among the invasive isolates in Switzerland and France, and biotype gravis dominated among isolates from local infections in Italy and the United Kingdom (5,8–10). All but 1 isolate from Poland represent a single clone despite isolation of the strains in different part of the country over a 9-year period. This phenomenon has not been documented in any countries reporting nontoxigenic C. diphtheriae infections. This raises a valid question: is a single clone of C. diphtheriae circulating in Poland or does the identified clone have increased pathogenic properties? This question remains unanswered because the carrier state of C. diphtheriae has not been examined in the Polish population.

Taking these data and the literature review into consideration, C. diphtheriae infections frequently are associated with endocarditis. Muttaiyah et al. (11) and Mishra et al. (12) demonstrated that most patients with C. diphtheriae endocarditis have underlying cardiac disease, prosthetic valves, or a history of IVDU. This finding, however, was not observed among patients in Poland.

The portal of entry for invasive nontoxigenic C. diphtheriae infection has not been fully elucidated. However, some authors shown that skin lesions are the most likely sources (9,13,14). In the cases presented here, skin ulceration was uncommon (1 case), but dental caries were found in >22% of cases. Dental caries could be a portal of entry.

The main limitation of this work is lack of complete data. Nevertheless, nontoxigenic C. diphtheriae can be concluded to be an emerging pathogen in Poland and has the potential to cause serious infections. The number of nontoxigenic C. diphtheriae infections might be higher because reporting of only toxigenic C. diphtheriae infections is mandatory in Poland. Moreover, in clinical settings, detection of coryneform bacteria in blood cultures is often dismissed as contamination, and in severe cases of the disease, C. diphtheriae might never be identified as the etiologic agent of bloodstream infection.

Homelessness, alcohol abuse, IVDU, and diabetes mellitus were mentioned in the literature as risk factors for C. diphtheriae invasive infections. In the cases presented here, 31% of patients were homeless, and 22% reported alcohol dependency but only 1 patient had diabetes mellitus. No patients reported IVDU. In 17% of cases, hepatic cirrhosis was ascertained, which suggests that it also may be another predisposing factor to infection. Moreover, dental caries is a highly probable portal of entry of C. diphtheriae invasive infection and has not been documented by other authors. However, such infections also might occur in persons with no identified predisposing factors.