Yersinia pseudotuberculosis and Y. enterocolitica Infections, FoodNet, 1996–2007

To the Editor: Yersinia pseudotuberculosis, a gram-negative zoonotic bacterial pathogen, causes acute gastroenteritis and mesenteric lymphadenitis, which are often accompanied by fever and abdominal pain. Although Y. pseudotuberculosis infections are distributed worldwide, little is known about their incidence and epidemiology in the United States. Y. pseudotuberculosis was first reported in the United States in 1938 and has rarely been identified since then (1). No outbreaks have been reported, and only 14 cases were documented from 1938 through 1973 (2). Although not reportable nationally, yersiniosis is a notifiable disease in all Foodborne Diseases Active Surveillance Network (FoodNet) sites. We describe the Y. pseudotuberculosis infections reported through FoodNet surveillance sites and compare these infections with those caused by the more commonly identified Yersinia species, Y. enterocolitica. 
 
During 1996–2007, FoodNet conducted active surveillance for laboratory-confirmed Yersinia spp. infections (excluding Y. pestis) in Connecticut, Georgia, Maryland, Minnesota, New Mexico, Oregon, Tennessee, and selected counties in California, Colorado, and New York. All clinical laboratories in these areas were routinely contacted to ascertain cases. Demographic and outcome (e.g., hospitalization and death) information was collected for all cases. On the basis of the source of specimen collection, infections were categorized as invasive (isolated from cerebrospinal fluid, blood, or another normally sterile site) or noninvasive (isolated from urine, stool, or other site). Data were analyzed by using SAS version 9.2 (SAS Institute, Cary, NC, USA). Differences were evaluated by using χ2 and Fisher exact tests, and medians were compared by using the Wilcoxon rank-sum test. A 2-tailed p value of <0.05 was considered significant. 
 
During 1996–2007, 1,903 Yersinia infections were reported in FoodNet sites. Of these, 1,471 (77%) had species information available. Most of the isolates were Y. enterocolitica (1,355; 92%); 18 (1%) Y. pseudotuberculosis infections were identified. The average annual incidence of Y. pseudotuberculosis infections was 0.04 cases per 1,000,000 persons. Most Y. pseudotuberculosis cases were reported from the western FoodNet areas of California (5 cases) and Oregon (5 cases). 
 
The median age of persons with Y. pseudotuberculosis infection was 47 years (range 16–86 years), and 67% were male (Table). Of the 13 Y. pseudotuberculosis cases for which race was reported, 10 (77%) were in whites. Eight (44%) Y. pseudotuberculosis cases occurred in the winter months (December–February). Thirteen (72%) persons with Y. pseudotuberculosis infection required hospitalization; the median hospital stay was 9 days (range 2–35 days). Two deaths were reported, yielding a case-fatality rate of 11%. Twelve (67%) of the Y. pseudotuberculosis isolates were recovered from blood specimens, and only 1 isolate was recovered from stool. 
 
 
 
Table 
 
Comparison of Yersinia pseudotuberculosis and Y. enterocolitica infections, FoodNet, 1996–2007* 
 
 
 
In comparison, the average annual incidence of Y. enterocolitica infections in FoodNet was 3.5 cases per 1,000,000 persons, and many of the cases occurred in the southern FoodNet site of Georgia (443 cases, 33%) (Table). Persons with Y. enterocolitica infection were significantly younger than those with Y. pseudotuberculosis infection (median age 6 years, p = 0.0002), and unlike Y. pseudotuberculosis infections, Y. enterocolitica infections were evenly distributed among male and female patients and among whites and blacks. Compared with those with Y. enterocolitica infection, persons with Y. pseudotuberculosis infection were more likely to be hospitalized (p = 0.0003), have longer hospital stays (p = 0.0118), die (p = 0. 0248), and have an isolate recovered from an invasive site (p<0.0001). 
 
Most of the Y. pseudotuberculosis infections reported in FoodNet sites appeared to be severe and invasive. The rarity of diagnosed Y. pseudotuberculosis infections is consistent with earlier reports from North America (3,4), but this rarity remains unexplained. This rarity contrasts with the observation that cases and outbreaks are more common in other parts of the developed world, particularly in northern climes (1,5,6–8). The recent appearance of epizootic Y. pseudotuberculosis in farmed deer in the southern United States suggests that this could change (9). 
 
The high proportion of Y. pseudotuberculosis cases that were diagnosed by blood culture suggests that less invasive Y. pseudotuberculosis infections are underrecognized in the United States. Diagnosis of Yersinia infections is difficult without specific culture, Yersinia is not routinely tested for in the United States, and isolation of the organism by culture may be difficult with standard media (2,10). Clinical diagnosis of Y. pseudotuberculosis infections can be challenging because physicians are not aware that Y. pseudotuberculosis is a potential cause of gastroenteritis (10). In the syndrome of pseudoappendicitis, the distinctive findings found by surgical exploration of severe mesenteric lymphadenitis can be suggestive, but diagnosis would require confirmation by culture of nodes or feces (2,3). 
 
Unless the physician is both aware of Y. pseudotuberculosis as a cause of gastroenteritis and knows which diagnostic test to order, Y. pseudotuberculosis infections will go undiagnosed. Clinicians should consider Y. pseudotuberculosis as a cause of gastroenteritis and pseudoappendicitis and request appropriate microbiologic testing for patients with suspected cases. If more cases are identified in the United States, another investigation of Y. pseudotuberculosis might clarify the epidemiology of this infection.

California, Colorado, and New York. All clinical laboratories in these areas were routinely contacted to ascertain cases. Demographic and outcome (e.g., hospitalization and death) information was collected for all cases. On the basis of the source of specimen collection, infections were categorized as invasive (isolated from cerebrospinal fl uid, blood, or another normally sterile site) or noninvasive (isolated from urine, stool, or other site). Data were analyzed by using SAS version 9.2 (SAS Institute, Cary, NC, USA). Differences were evaluated by using χ 2 and Fisher exact tests, and medians were compared by using the Wilcoxon rank-sum test. A 2-tailed p value of <0.05 was considered signifi cant.
During 1996-2007, 1,903 Yersinia infections were reported in FoodNet sites. Of these, 1,471 (77%) had species information available. Most of the isolates were Y. enterocolitica (1,355; 92%); 18 (1%) Y. pseudotuberculosis infections were identifi ed. The average annual incidence of Y. pseudotuberculosis infections was 0.04 cases per 1,000,000 persons. Most Y. pseudotuberculosis cases were reported from the western FoodNet areas of California (5 cases) and Oregon (5 cases).
The median age of persons with Y. pseudotuberculosis infection was 47 years (range 16-86 years), and 67% were male (Table). Of the 13 Y. pseudotuberculosis cases for which race was reported, 10 (77%) were in whites. Eight (44%) Y. pseudotuberculosis cases occurred in the winter months (December-February). Thirteen (72%) persons with Y. pseudotuberculosis infection required hospitalization; the median hospital stay was 9 days (range 2-35 days). Two deaths were reported, yielding a case-fatality rate of 11%. Twelve (67%) of the Y. pseudotuberculosis isolates were recovered from blood specimens, and only 1 isolate was recovered from stool.
In comparison, the average annual incidence of Y. enterocolitica infections in FoodNet was 3.5 cases per 1,000,000 persons, and many of the cases occurred in the southern Food-Net site of Georgia (443 cases, 33%) (Table). Persons with Y. enterocolitica infection were signifi cantly younger than those with Y. pseudotuberculosis infection (median age 6 years, p = 0.0002), and unlike Y. pseudotuberculosis infections, Y. enterocolitica infections were evenly distributed among male and female patients and among whites and blacks. Compared with those with Y. enterocolitica infection, persons with Y. pseudotuberculosis infection were more likely to be hospitalized (p = 0.0003), have longer hospital stays (p = 0.0118), die (p = 0.0248), and have an isolate recovered from an invasive site (p<0.0001).
Most of the Y. pseudotuberculosis infections reported in FoodNet sites appeared to be severe and invasive. The rarity of diagnosed Y. pseudotuberculosis infections is consistent with earlier reports from North America (3,4)  this rarity remains unexplained. This rarity contrasts with the observation that cases and outbreaks are more common in other parts of the developed world, particularly in northern climes (1,5,(6)(7)(8). The recent appearance of epizootic Y. pseudotuberculosis in farmed deer in the southern United States suggests that this could change (9). The high proportion of Y. pseudotuberculosis cases that were diagnosed by blood culture suggests that less invasive Y. pseudotuberculosis infections are underrecognized in the United States. Diagnosis of Yersinia infections is diffi cult without specifi c culture, Yersinia is not routinely tested for in the United States, and isolation of the organism by culture may be diffi cult with standard media (2,10). Clinical diagnosis of Y. pseudotuberculosis infections can be challenging because physicians are not aware that Y. pseudotuberculosis is a potential cause of gastroenteritis (10). In the syndrome of pseudoappendicitis, the distinctive fi ndings found by surgical exploration of severe mesenteric lymphadenitis can be suggestive, but diagnosis would require confi rmation by culture of nodes or feces (2,3).
Unless the physician is both aware of Y. pseudotuberculosis as a cause of gastroenteritis and knows which diagnostic test to order, Y. pseudotuberculosis infections will go undiagnosed. Clinicians should consider Y. pseudotuberculosis as a cause of gastroenteritis and pseudoappendicitis and request appropriate microbiologic testing for patients with suspected cases. If more cases are identifi ed in the United States, another investigation of Y. pseudotuberculosis might clarify the epidemiology of this infection.

Measles Outbreak, the Netherlands, 2008
To the Editor: From June 1 through October 16, 2008, an outbreak of 99 reported measles cases occurred in the Netherlands (1). This outbreak was the largest measles outbreak in the Netherlands since 1999-2000, when >3,200 cases, including 3 deaths, were reported (2).
In the Netherlands, clinical symptoms compatible with measles in a person with laboratory-confi rmed measles virus infection or an epidemiologic link to a laboratory-confi rmed case are notifi able (i.e., must be reported to public health authorities). The National Measles Reference Laboratory conducts genotyping and submits sequences to the World Health Organization European Region Measles Nucleotide Surveillance database (www. hpa-bioinformatics.org.uk/Measles/ Public/Web_Front/main.php).
Of the 99 measles cases reported in the 2008 outbreak, 40 were laboratory confi rmed and 59 were notifi ed based on an epidemiologic link. The fi rst case-patient in the outbreak was a 6-year-old unvaccinated resident of The Hague who had not been abroad in the month before onset of illness.