Case-based diphtheria surveillance data from each of 25 Diphtheria Surveillance Network (DIPNET) member countries (Austria, Belgium, Bulgaria, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Ireland, Italy, Latvia, Lithuania, the Netherlands, Norway, Poland, Portugal, Romania, Slovenia, Spain, Sweden, Turkey, and the United Kingdom) for 2000–2007 were submitted retrospectively to the coordinating center in the United Kingdom during 2008. Data for 2008 and 2009 were obtained in August 2009 and September 2010 from the DIPNET online database, which was launched in September 2007.

We analyzed cases meeting the DIPNET case definition (isolation of a toxigenic strain or clinically compatible case with an epidemiologic link to a laboratory-confirmed case) (Technical Appendix 2). In addition, 48 cases without laboratory confirmation and pseudomembrane (mild diphtheria/severe pharyngitis) and 5 cases with unknown manifestations were included for Latvia because these cases had been recorded in the national dataset. For most cases, toxigenicity was confirmed by using the Elek phenotypic test (9). However, in some cases, toxigenicity was evaluated only by detection of the toxin gene with PCR. We assumed that all cases in this dataset were toxigenic (toxin producing) because the number of cases without Elek confirmation was small and referred to symptomatic cases. Data fields collected included year; organism; biovar; and patient age, sex, clinical manifestations, vaccination status, veterinary contact, risk group, and outcome. Further strain characterization (ribotyping) was available for a limited number of isolates as part of a screening study in 10 DIPNET countries (10).

Cases were assigned to 5 clinical manifestation groups. These groups were classic respiratory diphtheria with pseudomembrane (the most serious form of the disease); mild diphtheria/severe pharyngitis (respiratory symptoms without the pseudomembrane); cutaneous (toxigenic organism isolated from skin lesions); other (e.g., toxigenic organism isolated from blood); and asymptomatic (carriers of toxigenic organisms, usually contacts of a confirmed case-patient).

Figure 1

Figure 1. Diphtheria Surveillance Network (DIPNET) and World Health Organization (WHO) European Region countries. 1, Albania; 2, Andorra; 3, Armenia; 4, Austria; 5, Azerbaijan; 6, Belarus; 7, Belgium; 8, Bosnia and Herzegovina; 9,...

Additional information concerning countries in the WHO European Region that are not DIPNET member countries was provided by the WHO Regional Office for Europe. Twenty-five of 53 member states of the WHO European Region are members of DIPNET. WHO European Region countries (including DIPNET members) report total cases annually to the WHO Regional Office for Europe through the WHO/United Nations Children’s Fund Joint Reporting Form, which is the global annual data survey of WHO member states for vaccine-preventable diseases and immunization program indicators. In addition, 16 countries in 2003 (Figure 1) were asked to prospectively participate in monthly surveillance and provide more detailed information (e.g., pathogen biovar; patient age, sex, and outcome; and carriers among contacts). Twelve countries currently provide monthly reports to WHO Regional Office for Europe through this system. The only major source of cases that has not participated in the monthly reporting system (but does report annually) is the Russian Federation. Rates per 1 million person-years were calculated by using population estimates derived from the Population Division of Economic and Social Affairs of the United Nations Secretariat (11).

Figure 2

Figure 2. Diphtheria cases per 1 million population in the World Health Organization (WHO) European Region and number of countries with a rate >1 cases/1 million population, 2000–2009.

Overall, across the WHO European Region, the number of cases of diphtheria has substantially decreased since the epidemic in the 1990s (Figure 2). Data on clinically confirmed cases and toxigenic isolates of C. diphtheriae and C. ulcerans reported to DIPNET during 2000–2009 are shown in Table 1 and Table 2, respectively. Member countries that are not listed reported no isolates. Data are analyzed separately for Latvia, where diphtheria is endemic.

Diphtheria-Endemic Countries in WHO European Region

During 2000–2009, Latvia reported the highest annual incidence rate of diphtheria in the European Region each year and a 10-year incidence rate of 23.8 cases/1 million person-years. This rate was ≈7× higher than in countries with the next highest 10-year incidence: i.e., Georgia (3.5), Ukraine (3.3), and the Russian Federation (3.0). However, during this time, 4,304 (>61%) of 7,032 cases in the WHO European Region were reported from the Russian Federation, and 2 countries, the Russian Federation and Ukraine, accounted for 83% of all cases.

Over the past 10 years, diphtheria incidence decreased by >95% across the region (from 1.82/1 million population in 2000 to 0.07/million in 2009), including in Latvia (from 111.22/million in 2000 to 2.67/million in 2009). In 2009, Latvia was the only country in the region that had not yet achieved the elimination benchmark of an incidence <1 case/million population (Figure 2).

Most cases reported to WHO through the monthly surveillance system were in teenagers and adults. However, the major risk groups for death have been infants (too young for complete primary vaccination) and adults >40 years of age (unvaccinated or with waning immunity). Although risk did not differ by sex in cases in children, during 2002–2009, ≈2× as many cases were reported in women >20 years of age than in men (510 [64%] vs. 292 [36%], respectively). Most (75%) case-patients reported in the European Region were at least partially vaccinated, but most (74%) case-patients and (93%) infants who died were unvaccinated). C. diphtheriae biovar gravis was the predominant strain (60%–80%). Of isolates from Latvia (Table 1), 355 (99%) of 358 with a known biovar were gravis and 3 (1%) were mitis.

Clinical manifestations and vaccination status for cases from Latvia (all C. diphtheriae) reported to DIPNET are shown in Table 3. Most (340/341) case-patients with symptoms had respiratory manifestations, and 141 (41%) of 340 respiratory case-patients had classic diphtheria symptoms. Vaccination showed a significant protective effect with respect to severity of infection (p<0.001 by test for trend).

Figure 3

Figure 3. Diphtheria incidence per 1 million person-years for Latvia (Corynebacterium diphtheriae, 2002–2009) and the remaining 24 Diphtheria Surveillance Network (DIPNET) countries (C. diphtheriae and C. ulcerans, 2000–2009). Error bars indicate 95% CIs....

For symptomatic cases for 2002–2009 (excluding the military outbreak in 2000 and cases from 2001 for which limited information was available) the highest overall incidences were in children 0–4 and 5–15 years of age and adults 45–64 years of age; lower incidence rates were observed in other age groups (Figure 3). Most (123/196, 63%) symptomatic cases during those years were in female patients.

The second most common risk factor (after military service) identified among symptomatic case-patients in Latvia was unemployment (60 case-patients). Information was not available regarding connections of case-patients to other countries of the former Soviet Union.

Substantial progress has been made in controlling diphtheria across Europe since the epidemic in the 1990s, but diphtheria has not disappeared as a serious public health threat. After major disruption to a mass vaccination program, recovery time is lengthy, and pockets of unvaccinated persons can remain because recovery is not necessarily homogeneous.

The protective effect of vaccination in preventing progression to severe disease is clear. However, 64 patients in Latvia recorded as fully vaccinated had classic respiratory diphtheria symptoms. Most of these patients were infected during a military outbreak in 2000 and would have been scheduled for primary vaccinations during the 1980s, when changes in vaccines, vaccination policy, medical practice, and public acceptance led to less intensive vaccination of children in the former Soviet Union. Beginning in 1980, Soviet vaccination recommendations enabled use of an alternative primary vaccination schedule against diphtheria that recommended 3 doses of a lower-potency vaccine (19). The classification of fully/partially vaccinated relies on specific interpretation of a country. Since the 2000 outbreak, greater attention has been given to checking vaccination records of new recruits into the Latvian military, and booster vaccinations are given where appropriate.

Lower CFRs for respiratory diphtheria in disease-endemic areas compared with those in nonendemic areas highlight how lack of familiarity with a rare disease can affect diagnosis and treatment. As the incidence of diphtheria has decreased, so has the practice of routine laboratory screening (20). No DIPNET member country routinely screens all throat swab specimens for corynebacteria, although sentinel screening of all throat swab specimens is conducted in Denmark, Ireland, and the United Kingdom. All other DIPNET countries (and outside sentinel screening areas) perform screening only at the request of the clinician or if the laboratory identifies particular criteria for screening from information accompanying a swab specimen (DIPNET, unpub. data). This practice has resulted in a loss of laboratory expertise and the opportunity for infections to go undetected because only clinically indicated swab specimens are tested; thus, milder cases or those with unusual manifestations may be missed.

A recent DIPNET external quality assurance evaluation of 6 simulated throat specimens found that only 6 of 34 international centers produced acceptable results for all 6 specimens; many centers could not isolate the target organism (21). In some poor countries, screening can be limited by cost of laboratory reagents, and problems have also occurred in obtaining Elek reagents and media (21). During a recent screening study across 10 countries in Europe, toxigenic organisms were isolated in Latvia and Lithuania (10). At least one of these cases in Lithuania would not have been correctly diagnosed in the absence of the screening study. In addition to the potential for missed or late diagnoses, in areas where diphtheria is not endemic, diphtheria antitoxin treatment is not always available, which can have serious consequences. A recent international survey highlighted global shortages of diphtheria antitoxin (22). Information about administration and timing of antitoxin treatment was not collected for this analysis, but studying such timing in relation to differing CFRs would be useful.

Higher incidence rates of C. diphtheriae among women in disease-endemic countries could be caused by several factors. Women more commonly work as caregivers in domestic and health care settings, consultation rates are usually higher among women, and men are more likely to have received diphtheria vaccine during military service.

Although the United Kingdom, France, and Germany regularly report isolations of toxigenic C. ulcerans, it is unlikely that this organism is present only in these countries. The ability to detect C. ulcerans could indicate the capability of a country to detect potentially toxigenic organisms and provide an indicator of good surveillance. Detection of mild diphtheria cases (any toxigenic organism) is another potential indicator of good surveillance. C. ulcerans appears to have a wide host range and has been isolated from many domestic and wild animals, including the killer whale and lion (nontoxigenic strain) (23). During 2002 and 2003, toxigenic C. ulcerans strains isolated from domestic cats in the United Kingdom were found to have the predominant ribotypes observed among human clinical isolates, which suggests that cats could be a potential reservoir for human infection (24). Identical C. ulcerans strains have been isolated from diphtheria patients and dogs in France and the United Kingdom (14,15). The presence of this organism reinforces the need to maintain high vaccination levels in all countries. Higher incidence of infection among elderly women could be related to pet ownership habits, in combination with low or waning immunity.

Vaccination coverage for diphtheria is assessed annually in many countries in Europe by using a range of methods, including computerized vaccination registers, survey methods, administrative methods, or a combination (25). These methods will provide varying degrees of accuracy in coverage estimates, which makes countries difficult to compare. Coverage for vaccination with diphtheria-tetanus-pertussis 3 vaccine (third dose of diphtheria, tetanus, pertussis vaccine) in early childhood in 2009 was >90% for most (85%) countries in the European Region, and 66% of countries (including Latvia, Lithuania, Turkmenistan, and the Russian Federation) reported coverage >95% (26). Coverage in Ukraine decreased from 98% in 2006 and 2007 to 90% in 2008 and 2009. Austria, Denmark, Georgia, and Moldova recorded diphtheria-tetanus-pertussis 3 vaccine coverage <90%. Azerbaijan and Malta had the lowest coverage (73% for both countries) in the European Region in 2009.

Following high-profile vaccine-scare stories in some countries in eastern Europe, such as the Russian Federation and Ukraine, anti-vaccination groups have gained strength by using television, the Internet, and other media for publicity (27); this activity could seriously affect vaccination coverage. Adult diphtheria immunity can be increased through scheduled booster vaccinations every 10 years (e.g., as in Austria, Belgium, Bulgaria, Cyprus, Estonia, Finland, France, Germany, Greece, Latvia, Norway, Portugal, and Romania) or as part of a combined tetanus and low-dose diphtheria vaccine given for tetanus-prone injuries. In Latvia, annual adult vaccination coverage surveys are undertaken, but in most countries adult coverage is rarely assessed. Seroprevalence studies have indicated that many adults in some countries have immunity levels below the protective threshold (28). Gaps in immunity in the adult population contributed to the resurgence of diphtheria in eastern Europe during the 1990s.

Trends in diphtheria cases in Europe are encouraging, but continued striving for improved vaccination coverage is essential. Diphtheria has a socioeconomic component; outbreaks are typically seen in marginalized groups. In the current economic climate, more socially deprived groups that are vulnerable to infection will emerge. The economic crisis may also threaten supplies of vaccine and antitoxin and delivery of immunization programs. Because reductions in finances can limit the capacity for surveillance, decreases in case reporting need to be interpreted with caution. Every effort must be made to maintain high diphtheria vaccination coverage.