Differences in Clinical and Laboratory Findings between Group D and Non-Group D Non-Typhoidal Salmonella Gastroenteritis in Children.

PURPOSE: To investigate the differences in clinical features and laboratory findings between group D and non-group D non-typhoidal Salmonella (NTS) gastroenteritis in children.
METHODS: A retrospective chart review of children diagnosed with NTS confirmed by culture study was performed. The clinical features and laboratory findings of group D and non-group D NTS were compared.
RESULTS: From 2003 to 2012, 75 cases were diagnosed as NTS at our center. The number of group D and non-group D patients was 45 and 30, respectively. The mean age was higher in group D than in non-group D patients (5.1 years vs. 3.4 years, p=0.038). Headaches were more frequently observed (p=0.046) and hematochezia was less frequently observed (p=0.017) in group D than in non-group D NTS gastroenteritis patients. A positive Widal test result was observed in 53.3% of group D and 6.7% of non-group D NTS cases (O-titer, p=0.030; H-titer, p=0.039). There were no differences in white blood cell counts, level of C-reactive protein and rate of antimicrobial resistance between group D and non-group D cases.
CONCLUSION: The more severe clinical features such as headache, fever, and higher Widal titers were found to be indicative of group D NTS gastroenteritis. Additionally, group D NTS gastroenteritis was more commonly found in older patients. Therefore, old age, fever, headache, and a positive Widal test are more indicative of group D NTS than non-group D NTS gastroenteritis. Pathophysiological mechanisms may differ across serologic groups.

INTRODUCTION

Salmonella enterica causes diseases in humans and can be divided into two groups: human-restricted typhoidal Salmonella serovars (Typhi and Paratyphi) causing typhoid fever, and non-typhoidal Salmonella(NTS) serovars. NTS are well-known foodborne diseases including gastroenteritis, and bacteremia with subsequent focal infection. The most common NTS infection is self-limiting gastroenteritis in humans and animals [1]. The global burden of NTS gastroenteritis remains high at 93.8 million cases per year [2]. Control of Salmonella infection is difficult due to the bacterium's widespread distribution, multiple drug resistance, adaptability, and high tolerance to environmental stress [3].

Salmonella strains include over 50 serogroups based on the O antigen, and are classified into more than 2,500 serotypes that have a unique combination of somatic O and flagella H1 and H2 antigens [4]. Serotypes adapted to humans, such as S. typhi and S. paratyphi, cause enteric fever in humans. Among them, S. typhimurium (group B) and S. enteritidis (group D) cause infection in humans and animals [5]. Groups B, C and D NTS are most commonly isolated in children younger than 5 years of age [2].

Recently reported cases of group D NTS include a case with an infected pseudoaneurysm [6]; a case of enteric fever with bowel perforations caused by group D NTS [7]; and a colonic ulcer caused by group D NTS [8]. Group D has been reported as the most commonly isolated serogroup of NTS gastroenteritis in Korea [91011]. To the best of our knowledge, there has been no study to date that has compared group D NTS and non-group D NTS in Korean children. In the current study, we compared the clinical and laboratory data, obtained on the day of hospital admission, of group D and non-group D NTS gastroenteritis in children from 2003 to 2012.

MATERIALS AND METHODS

Patients and samples

We conducted a retrospective review of demographic, clinical, and laboratory data of all children, <16 years of age, hospitalized with laboratory-confirmed NTS at Gyeongsang National University Hospital in Jinju, Korea between January 2003 and December 2012. Cases of NTS were confirmed with stool or blood cultures. Salmonella strain serogroups, but not serovars, were identified with commercial agglutination tests.

Data collection

Basic demographic and clinical data included age, sex and date of admission; presence of fever, abdominal pain, headache and hematochezia; and antimicrobial agent treatment. Laboratory parameters included total white blood cell (WBC) counts, C-reactive protein (CRP) levels, Widal O- and H-titers, blood and stool cultures, and antimicrobial resistance. A Widal titer (O- and H-titers) of greater than 1:40 was considered positive. This study was approved by the Gyeongsang National University Hospital institutional review board following their review of the research protocols (GNUHIRB-2013-02-005).

Bacteria and antimicrobial resistance test

McConkey and Salmonella-Shigella agar plates were used. The sample serotype was determined using the serotype agglutination test from the Korea Centers for Disease Control and Prevention. Antimicrobial resistance testing was performed with the VITEK 2 system (Bio-Merieux, Durham, NC, USA). Interpretive criteria as outlined by the Clinical and Laboratory Standards Institute were applied.

Statistics

Data from group D and non-group D patients were compared using a Student t-test for normally distributed data, and differences in population proportions were analyzed using the χ2 test. A p-value of less than 0.05 was considered statistically significant. IBM SPSS Statistics ver. 21.0 for Windows (IBM Co., Armonk, NY, USA) was used for statistical analysis.

RESULTS

A total of 83 children, <16 years of age, with culture-positive Salmonella infections were admitted to Gyeongsang National University Hospital from January 2003 to December 2012. Among them, 8 patients were diagnosed with typhoid fever and 75 patients were diagnosed with NTS gastroenteritis. Of these 75 cases, 74 were isolated from stool and 1 was isolated from blood. All patients were managed with oral or intravenous antimicrobial agents (third-generation cephalosporins), which are used to treat either typhoid fever or bacterial gastroenteritis such as shigellosis.

The incidence rate of NTS gastroenteritis increased from 2003 to 2012 (p=0.033, Fig. 1). The lowest incidence rate was 1, in 2003, and the highest was 11, in 2011. The leading Salmonella serogroup was group D (60.0%), followed by group B (25.3%), group C (9.3%), and group E (5.3%). The monthly incidence rate of NTS gastroenteritis was higher from May to October than from November to April (p=0.007, Fig. 2).

Fig. 1

Annual incidence rates of non-typhoidal Salmonella (NTS) gastroenteritis. The incidence rate of non-typhoidal salmonellosis increased from 2003 to 2012 (p=0.033). The most commonly isolated NTS serogroup was group D. Group D was the most isolated and then were isolated by group B, group C, group E.

Fig. 2

Monthly Incidence rate of non-typhoidal salmonellosis. The incidence rate was highest in May and was higher from July to October compared with other months (p=0.007).

The population age proportions of <1 year, 1-5 years, 6-10 years, and 11-15 years were 12.0% (9/75), 60.0% (45/75), 13.3% (10/75) and 14.7% (11/75), respectively (Fig. 3). The mean age was 4.4 years (range: 0-15 years). The male-to-female ratio was 1:1 (49.3% and 50.7%). None of the patients had a previous history of chronic disease or malignancy before the episode of NTS gastroenteritis. All patients had diarrhea and 94.7% of patients had fever. Other symptoms included abdominal pain (57.0%), hematochezia (25.3%) and headache (17.3%). There was no pneumonia, arthritis, or meningitis present in any patient. All patients were managed with oral or intravenous antimicrobial agents (third-generation cephalosporins) and discharged in good general condition.

Fig. 3

Number of group D and non-group D non-typhoidal Salmonella gastroenteritis patients according to age. The peak occurrence was observed in children aged 2 and 3 years old in both groups. In children under the age of 5, there were more non-group D cases than group D cases.

The median WBC count was 10,190/mm3 (range: 4,190-24,970/mm3). The median CRP level was 48.7mg/L (range: 0-284.6 mg/L). High CRP levels (>100 mg/L) were found in 17 patients and normal CRP levels (<5 mg/L) were found in 3 patients. The Widal test was performed in 47 patients with high fever and/or abdominal pain. A positive Widal titer (O- or H-titer >1:40) was observed in 24 patients. An O- or H-titer >1:160 was observed in 15 patients.

Antimicrobial susceptibility tests detected resistance to ampicillin (41.3%), cefazolin (33.3%), cefotaxime (6.7%), ceftazidime (6.7%), gentamicin (13.3%), tobramycin (17.35), piperacillin (29.3%), and trimethoprim/sulfamethoxazole (2.7%) (Table 1). All isolates were susceptible to ciprofloxacin. The tobramycin-resistant strain was absent from 2003 to 2007 but was present from 2008 (Table 1).

Table 1

Resistance Rate to Antimicrobial Agents according to the Year of Isolation

NA: not available.

Comparison of clinical features and laboratory findings between group D and non-group D NTS gastroenteritis

The number of patients diagnosed with group D and non-group D NTS was 45 and 30, respectively. The mean age was higher in group D (5.1 years) than in non-group D cases (3.4 years) (p=0.038, Table 2). The male to female ratio was 1.5:1 in group D and 1:2 in non-group D cases, respectively; it was statistically significant (p=0.021). Accompanying headaches were present in 24.4% of group D and 6.7% of non-group D cases (p=0.046). Hematochezia was more frequently found in non-group D (40.0%) than in group D cases (8.9%) (p=0.017). No differences were observed in frequency of fever, length of hospitalization, WBC counts, CRP levels, and resistant rates of antimicrobial agent.

Table 2

Comparison of Clinical and Laboratory Findings between Group D and Non-Group D Non-Typhoidal Salmonella Gastroenteritis

Values are presented as number only, mean±standard deviation, or number (%).

The Widal titer was higher in group D than in non-group D cases (O-titer, p=0.030; H-titer, p=0.039). O titers and H titers of >1:640 were observed in 8 patients; all were detected in group D NTS cases (Table 3). The Widal titer was not correlated with fever, headache, hematochezia, WBC count, CRP level, or antimicrobial resistance.

Table 3

Clinical and Laboratory Findings of Group D Non-Typhoidal Salmonella Gastroenteritis Pateints with High O- or H-titers (≥1:640)

WBC: white blood cell, CRP: C-reactive protein, R: resistance.

DISCUSSION

During the study period from 2003 to 2013, group D salmonella was the predominant gastroenteritis-causing NTS serogroup identified in children in the Jinju area of Korea. Clinical features of group D NTS were similar to those of typhoid fever; these features included fever, headache, abdominal pain and a positive Widal test result. Compared with non-group D NTS patients, the mean age was higher and headaches were more frequently observed in group D patients. A positive Widal test result was observed in 53.3% of group D cases in the present study.

In 2008, NTS gastroenteritis accounted for 34% of NTS cases in children in Korea [12]. Since then, the incidence rate has increased yearly: 42.1% in 2009; 52.8% in 2010; and 63.7% in 2011 [131415]. Table 4 shows a summary of recent Korean pediatric NTS cases, including those in our study. In the present study, we found that the number of children with NTS gastroenteritis increased from 2003 to 2012. However, the incidence of NTS decreased from 141 during 1986-1995 [9] to 75 during 2003-2012; one possible explanation for this is that different study areas were investigated during the different periods (Table 4) [91011].

Table 4

Summary of the Reported Non-Typhoidal Salmonella Gastroenteritis (NTS) Cases in Children in Korea

NA: not available, TMP-SMZ: trimethoprim-sulfamethoxazole.

In the studies of Korean pediatric NTS cases [91011] (Table 4), group D was the most commonly isolated NTS serogroup, accounting for 62.5% of cases during 1998-2008 [10] and 46.0% of cases during 2000-2011 [11]. However, during 1986-1995, group D NTS was only 17.7%, and group B (57.4%) was the most commonly isolated in children with NTS [9]. Although the different research areas may account for this change, there may also have been a change in salmonella strain in Korea from group B to D.

In infants, an inoculum of less than 106-108 NTS organisms can cause disease. Direct person-to-person transmission, although uncommon, sometimes occurs. NTS infection especially occurs in children under 2 years of age [16] and the incidence rate in children younger than 1 year of age is 10 times higher than in the general population [17]. NTS infection in Korea has been previously reported to be more prevalent in children younger than 5 years of age than in children aged between 5 and 15, regardless of serogroup [91011]. This is similar to our finding that 72% of NTS infections occurred in children under the age of 5. In western countries, the highest incidence of salmonellosis has been reported to occur in infants under the age of 1 year [1819]. However, in present study, the percentage of patients who were infants (<1 year old) was 12.0%, lower than the percentage of children aged 1 and 2 years (17.3% each). However, the mean age of children with NTS did not differ between study areas and periods (Table 4). For patients older than 1 year of age, the present study showed that the number of infections caused by group D Salmonella exceeded that of non-group D Salmonella. This finding differs from a previous report that found that from 2001 to 2011, Taiwanese children younger than 5 years of age were more likely to have group B infection [3]. One possible explanation for this is the difference in epidemic Salmonella serogroups found in the different countries. The different age groups of affected children might be associated with different lifestyles, diets, and environmental conditions such as attending a daycare center, eating undercooked eggs, and contact with cats and reptiles [2021].

In 1989, Salmonella infection occurred throughout the year and this was attributed to its indigenization in Korea at this time [22]. A recent study of 2012-2013 cases showed that the peak season for Salmonella infection was autumn (September to November) [2324]. In present study, the seasonality of NTS gastroenteritis was summer and autumn (May to October); this was similar to the reports of 1986-1995 cases in Seoul [9] and of 2000-2011 cases in Seongnam [10]. However, in Jeonju, the peak incidence of NTS was observed from April to June during 1998-2008 [11] (Table 4). The difference in peak season might be due to regional differences.

The clinical symptoms of NTS gastroenteritis are acute onset of fever and chills; nausea and vomiting; and abdominal cramping and diarrhea [25]. In the present study, the common symptoms of NTS were diarrhea, fever, and abdominal pain. Hematochezia and headache were also observed in some patients. Hematochezia is a relatively common presentation in NTS gastroenteritis [31011], but headache is not. In the present study, headaches were more frequently observed in group D NTS than in the non-group D gastroenteritis cases, while the reverse was true for hematochezia. Compared with other reports of Korean children with NTS, hematochezia was less frequently, and fever was more frequently observed in our study (Table 4). Therefore, in the present study, for cases with positive Widal test results, typhoid fever was considered the first possible diagnosis.

Recently, invasive NTS infection has emerged in children [262728]; group D is the most common serogroup in these infections [29]. Bacteremia was diagnosed in a 1-year-old child in this study and group D NTS was isolated.

Although the Widal test is usually used for the diagnosis of typhoid fever caused by S. typhi [3031], half of the patients (51.1%, 24/47) who were checked tested positive, and a positive Widal test was found in 53.3% (24/45) of group D NTS gastroenteritis cases. The positive Widal titers could be associated with invasive NTS infection, including bacteremia, because elevated CRP levels were also observed with the positive Widal test results.

Antimicrobial agents are not generally recommended for the treatment of isolated uncomplicated NTS gastroenteritis because they may suppress normal intestinal flora and prolong both the excretion of Salmonella and the remote risk of creating the chronic carrier state [32]. However, some studies have suggested that antimicrobial treatment may be associated with an improvement in symptoms and a more rapid clinical recovery [33]. Studies of empirical antimicrobial therapy (before culture results are back) in severe community-acquired diarrhea have also found a reduction in disease duration of one to two days [3435]. Additionally, patients with high CRP levels (≥100 mg/L) are more frequently administered empirical antimicrobial therapy and experience more complications than others [36]. Therefore, until the culture results can confirm fever and leukocytosis, CRP levels should be checked prior to the administration of the antimicrobial agent. In the present study, all patients were managed with an antimicrobial agent because most of the patients had general symptoms, including fever or headache and high CRP levels. This was a retrospective study and we could not find specific reasons for the antimicrobial use.

Antimicrobial resistance in NTS is an important problem, although most NTS cases do not require antimicrobial treatment. Resistance is associated with an increased risk of infection, hospitalization, and death [3738]. Surveillance data shows an obvious increase in overall antimicrobial resistance among salmonellae from 20% to 30% in the early 1990s to as high as 70% in some countries at turn of the century [39]. In our study, antimicrobial resistance to one or more antimicrobial agents was 66.7% and this was similar to other reports [28]. The resistance to ampicillin (41.3%), third-generation cephalosporins (6.7%), trimethoprim/sulfamethoxazole (2.7%) and quinolone (0%) was lower compared with children in Thailand [28], Taiwan [29] and India [40]. However, in Korea the resistance to ceftriaxone was higher in our study (6.7%) than in other studies (Table 4).

In summary, since 2000 the incident rate of NTS, as investigated by our study, was not different to that of other areas in Korea despite the increased number of reported cases from 2003 to 2012. Group D was the most commonly isolated Salmonella serogroup in our study and this was similar to the other recent reports in Korea. The majority of the clinical and laboratory findings were same between group D and non-group D NTS. However, headache and a positive Widal test result were more frequently observed in group D than non-group D cases. More severe clinical features such as headache, fever and higher Widal titers can be indicative of group D NTS gastroenteritis. Additionally, group D NTS gastroenteritis was more common in older children. Therefore older age, fever, headache and a positive Widal test result suggests group D NTS gastroenteritis more than non-group D NTS and, depending on the serotype, pathophysiologic mechanisms may differ by serologic groups.

There were some limitations in the present study. First, it was a retrospective study conducted in a single center. Second, genetic studies or serotyping of Salmonella strains were not performed because NTS was not considered an important disease in a normal healthy population. Therefore, no further genetic studies were undertaken. We believe that further studies of the changing patterns of the clinical features and laboratory findings of group D NTS in children are needed.