SEARCH

SEARCH BY CITATION

Keywords:

  • Bacilli;
  • climatic factors;
  • Gram-negative;
  • nosocomial;
  • seasonality

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Seasonality of Infections Caused by Gram-negative Bacilli
  7. HCAIs
  8. Discussion
  9. Transparency Declaration
  10. References

Clin Microbiol Infect 2012; 18: E934–E940

Abstract

To assess seasonal variations in Gram-negative and healthcare-associated infections (HCAIs), a literature search was performed with combinations of the keywords ‘seasonality’, ‘seasonal variations’, ‘Gram-negative bacilli’, ‘infections’, ‘nosocomial infections’, and ‘health care associated infections’, to retrieve articles published in English in peer-reviewed journals from 1 January 1970 to 29 February 2012. Seasonality was demonstrated for infections, mostly bloodstream infections (BSIs), caused by Acinetobacter spp., Escherichia coli, Enterobacter cloacae, Klebsiella spp., and Pseudomonas aeruginosa, with higher rates of infection during the summer months in North America, Europe, the Middle East, Australia, and Asia. Correlations were observed between temperature increase and rates of BSI for Acinetobacter spp., P. aeruginosa, E. coli, Klebsiella pneumoniae, and extended-spectrum β-lactamase-producing Enterobacteriaceae. A significant correlation between lower urinary tract infections and higher temperature and decreased relative humidity could explain the seasonality of some BSIs. Regarding HCAI, seasonality is intrinsically present in most viral respiratory and gastrointestinal infections, because viruses are introduced into hospitals during seasonal community outbreaks. Other HCAIs subject to seasonal variations include surgical wound infections, with winter peaks in the USA and summer peaks in Finland, central-line-associated BSIs in haematology/oncology paediatric outpatients, and dialysis-associated peritonitis. In summary, seasonal variations have been shown for infections caused by many Gram-negative bacilli, as well as for a few HCAIs, but many studies remain to be performed in order to better understand the mechanisms of these variations.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Seasonality of Infections Caused by Gram-negative Bacilli
  7. HCAIs
  8. Discussion
  9. Transparency Declaration
  10. References

Seasonality is defined as a periodic surge in disease incidence corresponding to seasons or other predefined calendar periods. Recognition of seasonal variations in the occurrence of infectious diseases dates back to the time of Hippocrates, who wrote in book III of the Aphorisms that ‘Every disease occurs at any season of the year but some of them more frequently occur and are of greater severity at certain times’. For Hippocrates, infections were influenced by environmental factors such as air, water, or food, and seasonal changes could give rise to diseases [1,2].

Later, identification of seasonal patterns during influenza epidemics made physicians wonder about the link to the winter season, and the name ‘influenza’ was used by Italians from the middle of the 18th century as influenza di freddo, or ‘influence of the cold’.

The first scientific approach to the study of seasonality in infections began during the 18th century, when the rise and fall of measles deaths were recorded and analysed [3].

As compared with community-acquired infections, for which seasonal variations have been identified, little is known about infections caused by Gram-negative bacilli and healthcare-associated infections (HCAIs). This review assesses whether or not seasonality has been described for infections caused by Gram-negative bacilli and HCAIs.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Seasonality of Infections Caused by Gram-negative Bacilli
  7. HCAIs
  8. Discussion
  9. Transparency Declaration
  10. References

A literature search was performed with different combinations of the keywords ‘seasonality’, ‘seasonal variations’, ‘Gram-negative bacilli’, ‘infections’, ‘nosocomial infections’ and ‘health care associated infections’ to retrieve articles written in English published in peer-review journals from 1 January 1970 to 29 February 2012.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Seasonality of Infections Caused by Gram-negative Bacilli
  7. HCAIs
  8. Discussion
  9. Transparency Declaration
  10. References

The Medline search retrieved 27 articles on the seasonality of infections caused by Gram-negative bacilli and HCAIs, including nine on bloodstream infections (BSIs), one on urinary tract infections (UTIs) and 16 on different types of HCAI in different settings. The heterogeneity of these studies makes their synthesis difficult.

Seasonality of Infections Caused by Gram-negative Bacilli

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Seasonality of Infections Caused by Gram-negative Bacilli
  7. HCAIs
  8. Discussion
  9. Transparency Declaration
  10. References

BSIs caused by Gram-negative bacilli

During the last 3 years, the results of several studies assessing the seasonality of BSIs have been published. Some studies included a wide range of organisms, whereas others focused on BSIs caused by selected microorganisms.

Study on multiple bacterial species.  Eber et al. have published the results of a study on inpatients from 132 US hospitals. Acinetobacter spp. infections exhibited the greatest seasonal variations, with a 52% increase in the summer as compared with the winter months, whereas the frequency of Escherichia coli BSIs increased by 12%. No seasonality was observed for Enterococcus spp. or Staphylococcus aureus. An increase in mean monthly temperature of 5.6°C corresponded to increases in BSI of 10.8% for Acinetobacter baumannii, 3.5% for E. coli, 8% for Klebsiella pneumoniae, 7.5% for Pseudomonas aeruginosa, and 2.2% for S. aureus. In time-series analysis, higher temperature was positively associated with BSI caused by Gram-negative bacilli and S. aureus in winter, spring, and autumn, whereas in summer, higher temperature was positively associated with BSI caused by Gram-negative bacilli, except for E. coli [4].

Gonçalvez-Pereira et al. assessed the impact of BSIs on admission to hospital on the outcome of 803 patients with community-acquired sepsis admitted to 17 intensive-care units. Gram-negative BSIs were significantly more common in the summer, whereas Gram-positive BSIs were more frequent in the winter [5].

E. coli BSIs.  Two studies on the seasonality of BSIs caused by E. coli were published in 2009 and 2010. Al-Hassan et al. from the Mayo Clinic performed a population-based investigation in Olmsted County (Minnesota). The overall incidence of E. coli BSIs was 41.4 per 100 000 person-years, 59% of the cases were community-acquired, and the urinary tract was the most common primary source of infection (80%). Significant seasonality was detected in the incidence of E. coli BSIs. The gender-adjusted and age-adjusted incidence rates of E. coli BSIs per 100 000 person-years were 50.2 during the warmest 4 months of the year (June–September) and 37.1 during the other 8 months, representing a 35% increase in incidence rate during the warmest 4 months. When the warmest 2 months were considered, there was a 44% increase in incidence in comparison with the other 10 months [6].

A hospital-based study was performed by Chazan et al. in northern Israel, by including 938 E. coli BSIs diagnosed during a 96-month period. Seventy-two per cent of the BSIs were community-acquired. The number of E. coli BSIs was significantly higher during the summer season than during either the winter or transitional seasons (31 ± 6 vs. 27 ± 4 vs. 28 ± 6). The incidence rates of E. coli BSIs were 19% and 21% higher in the summer than in the transitional and winter seasons, respectively, and temperature was positively correlated with the rate of E. coli BSIs, but no association between temperature and E. coli BSI incidence rates was found within seasons [7].

Klebsiella spp. BSIs.  Two studies assessing seasonal variation in K. pneumoniae BSIs published in 2009 and 2010 gave conflicting results.

The first study, by Anderson et al., took place on four different continents. The authors analysed surveillance data from 2001 to 2006 at four hospitals (Durham, NC, USA; Marseille, France; Melbourne, Australia; and Taipei, Taiwan). Incidence rates and incidence rate ratios were determined with multivariate Poisson regression. A total of 1189 K. pneumonia BSIs were included. The incidence rate of K. pneumoniae BSIs varied between institutions, but common features were identified, with the highest rates being observed in all institutions during the warmest 4 months of the year, the overall rate of BSIs being 1.5-fold higher during this period. Multivariable linear regression showed that temperature and dew point (relative humidity) were predictive of increased rates of K. pneumonia BSIs. Neither Enterobacter spp. nor Serratia spp. BSIs, used as control groups, showed seasonal variations [8].

In the second study, Al-Hasan et al. used a population-based study design identical to the one that they used for their study on E. coli BSIs. A total of 127 patients with Klebsiella spp. BSIs (K. pneumoniae, n = 105; Klebsiella oxytoca, n = 21; Klebsiella ornithinolytica, 1) diagnosed from 1998 to 2007 at the two laboratories of Olmsted county were included. Half of the cases were HCAIs, and the urinary tract was the most common primary source of infection (43%), followed by the gastrointestinal tract (24%) and the respiratory tract (10%); 20% were primary BSIs. A multivariate Poisson analysis adjusting for age, sex and average temperature did not show a linear temporal change in the incidence rate throughout the study period. In addition, no association was found between the incidence rate of Klebsiella spp. BSIs and average temperature [9].

Enterobacter spp. BSIs.  In 2011, Al-Hasan et al. performed a third study on the seasonality of BSIs by analysing Enterobacter spp. BSIs as they did for E. coli and K. pneumoniae BSIs. All Enterobacter spp. BSIs occurring in Olmsted county from 1998 to 2007 were included. They found, by multivariate Poisson regression, a significant linear trend of increasing incidence rate from 0.8 to 6.2 per 100 000 person-years between 1998 and 2007, but no significant difference in the incidence rate of Enterobacter spp. BSIs between the warmest 4 months and the remainder of the year [10].

UTIs

Falagas et al. examined whether lower UTIs are associated with meteorological parameters by retrospectively evaluating the correlation between the weekly percentage of house call visits for lower UTIs from 2000 to 2005 and the average weekly temperature and humidity recorded in the same area 3 days earlier. They included 3221 visits for lower UTIs in patients 62.9 ± 21.0 years of age. A significant correlation was found between visits for lower UTIs and the average higher weekly temperature and decreased relative humidity. Urine cultures were not systematically performed, and the microorganisms responsible for the UTIs are not listed in the article [11].

HCAIs

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Seasonality of Infections Caused by Gram-negative Bacilli
  7. HCAIs
  8. Discussion
  9. Transparency Declaration
  10. References

Enteric and respiratory viral infections

Both viral respiratory tract and gastrointestinal infections are intrinsically subject to seasonal variations, because viruses are introduced into healthcare facilities during seasonal community outbreaks when infected patients are admitted. Norovirus, the most common aetiological agent of acute gastroenteritis, is responsible for winter outbreaks in a wide spectrum of community and healthcare settings. Noroviruses alone are responsible for >50% of all reported outbreaks of HCAI [12].

In Germany, the results of an epidemiological study aimed at determining the proportion of HCAIs among all hospitalized cases of gastrointestinal infections from 2002 to 2008 showed that 39 424 (49%) of 80 650 norovirus infections and 11 592 (14%) of 83 451 rotavirus infections were healthcare-associated. Peaks of infections were observed in the winter for norovirus and in the winter and early spring for rotavirus [13].

In Edinburgh, Scotland, all hospitals and the community were actively monitored for outbreaks of gastroenteritis from September 2007 to June 2009. A total of 1732 patients and 599 healthcare workers were affected in 192 outbreaks. The number of outbreaks peaked in November and began to decrease in May [14].

HCAIs caused by Gram-negative bacteria

In 2008, Perencevich et al. published the results of a study whose objective was to identify potential seasonal trends in hospital infections. They used a cohort of all adult patients (218 594) admitted to the University of Maryland Medical Center from 1998 to 2005 with unique clinical cultures (26 624). Time-series analyses were used to estimate the association of the number of infections per month with season and temperature while controlling for long-term trends. During the summer months, increases in the mean rates of infection of 28% for P. aeruginosa, 46% for Enterobacter cloacae, 12% for E. coli and 21% for A. baumannii were identified. For each 5.6°C increase, a 17% increase in the monthly rate of infections caused by P. aeruginosa and A. baumannii was observed, and for the other pathogens higher temperatures were associated with higher infection rates, independently of seasonality [15].

Acinetobacter spp.

There is a need for a word of caution about the results of epidemiological studies on A. baumannii, because the quality of epidemiological data is questionable, owing to the difficulty in correctly identifying this organism. Nevertheless, Acinetobacter is one of the first organisms responsible for HCAI that has been shown to be subject to seasonal variation.

Two of the main studies on the seasonality of Acinetobacter spp. HCAIs have been performed by epidemiologists of the US CDC. The first study, performed by Retailliau et al., used the hospital-wide surveillance data of the National Nosocomial Infection Surveillance from 1974 to 1977, which showed an unexpected increase in HCAI rates during the late summer months. Seasonal variations were not associated with hospital geographical location or teaching status [16].

The second CDC study on the seasonality of Acinetobacter spp. HCAIs was performed by McDonald et al., who used 3347 Acinetobacter spp. infections reported to the National Nosocomial Infection Surveillance from 1987 to 1996. The authors found both a general downward trend in yearly infection rates for all infections, and an incidence rate of infections that was 54% higher during the July–October period than during the November–June period. A significant increase was observed for BSIs and pneumonia. The incidence of Acinetobacter spp. infections by geographical region ranged from 1.6 infections per 10 000 per patient-days in the west south-central region of the USA to 11.6 per 10 000 patient-days in the mid-Atlantic region. The percentage of incidence rate increase between the November–June period and the July–October period varied by geographical region, ranging from 14% in the west north-central states to 121% in the New England region. In contrast to Acinetobacter spp. infections, the monthly rate of P. aeruginosa infections varied little from the winter to summer months, with a 13% increase from the November–June period to the July–October period [17].

Gales et al. confirmed the CDC results with an epidemiological analysis of 30 374 strains of Acinetobacter spp. isolated from blood cultures obtained from patients from Canada, the USA and Latin America from 1997 to 1999 in the SENTRY Antimicrobial Surveillance Program. The highest number of Acinetobacter spp. infections occurred during the summer. However, in 1999, an opposite seasonal occurrence was observed, with a peak during the winter, influenced by an outbreak in a Brazilian hospital. Similarly, in the USA and Canada, peaks of infection during the winter, autumn and summer were also attributable to outbreaks [18].

An important question about the seasonality of infections concerns the behaviour of clones belonging to the same species. Are all strains subject to seasonal variations? Christie et al. studied the epidemiology of A. baumannii infections during a period of increased seasonal incidence by using molecular typing methods. At their hospital, from 1990 to 1992, the rate of A. baumannii isolation was significantly higher (30.4 per 1000 culture isolations) during summer than during autumn, winter, and spring (12.6 per 1000 culture isolations). Pulsed-field gel electrophoresis typing showed that 83% of 71 isolates had distinct restriction patterns. Three small clusters of isolates with the same pulsed-field gel electrophoresis patterns were identified, suggesting cross-transmission. However, the seasonal increased incidence of A. baumannii device-related HCAI in severely ill patients was not caused by clonal dissemination of a single strain, but by multiple strains [19].

Central-line-associated BSI (CLABSI)

Smith et al. investigated an increase in CLABSI rates among paediatric haematology-oncology outpatients by performing a case–control study comparing haematology-oncology outpatients with CLABSI (cases) with randomly selected haematology-oncology outpatients with a central venous catheter but no BSI during a 3-year period. They identified a significant increase in BSIs caused by ‘non-endogenous’ Gram-negative bacilli, such as Pseudomonas spp., during the summer months (May–October) as compared with the rest of the year [20].

Surgical wound infections

In Finnish folklore, the time from 12 June to 23 August has been called the ‘rotten month’ (dog days), because wound healing is delayed because of infections. A retrospective study was made of surgical infections entered in the infections register at Töölö hospital over the dog days from 2002 to 2005. Eight per cent of the 49.517 surgical operations were performed during the dog days, with a rate of surgical wound infections twice that at other times of the year [21].

Extended-spectrum β-lactamase (ESBL)-producing Enterobacteriaceae

Monthly data on ESBL-producing E. coli and Klebsiella spp. strains were collected retrospectively at two large university hospitals in Germany, and correlated with mean monthly temperatures by multivariable time-series analyses. Kaier et al. found both an increasing trend in the incidence of carriage of ESBL-producing bacteria and a positive correlation with mean temperature. More cases involving ESBL-producing bacteria were detected in the summer than in the winter [22].

Dialysis-associated peritonitis

Peritoneal dialysis represents 11% of all dialysis procedures performed worldwide, and peritonitis is the leading infectious complication [23]. Two studies have sought to assess the existence of seasonal variations in the incidence of dialysis-associated peritonitis. In the first one, by Kim et al., the incidence of continuous ambulatory peritoneal dialysis peritonitis according to temperature and relative humidity was analysed. During 1123 patient-months, 53 cases of peritonitis occurred, with the highest incidence in July and the lowest in November. A significant positive correlation was observed between the monthly frequency of continuous ambulatory peritoneal dialysis peritonitis, temperature, and relative humidity. Gram-negative peritonitis occurred uniformly throughout the year, whereas the rate of Gram-positive peritonitis increased during hot and humid months, with 50% of peritonitis cases occurring from March to August and 18% from September to February. Gram-negative organisms caused 7.3% and 29.4%, respectively, of peritonitis cases during the same periods [24].

In the second study, performed by Cho et al., all 6610 Australian patients receiving peritoneal dialysis between 2003 and 2008 were included to study the influence of seasons on peritonitis rates. When winter was used as the reference season, the peritonitis incidence rate ratios were 1.02 for summer, 1.01 for autumn, and 0.99 for spring. Significant seasonal variations were observed in the rates of peritonitis caused by coagulase-negative staphylococci (spring and summer peaks), corynebacteria (winter peak), and Gram-negative organisms (summer and autumn peaks). There were trends for seasonal variations in fungal peritonitis (summer and autumn peaks) and Pseudomonas peritonitis (summer peak). No significant seasonal variations were observed for other organisms. Peritonitis outcomes did not vary according to season [25].

HCAIs related to hospital water

Picard et al. described a total of 109 infections caused by Aeromonas hydrophila occurring at a hospital in the Paris area between 1973 and 1985. From 1983 to 1985, water samples were taken twice monthly for culture from the two main supply pipes entering the hospital and from three storage tanks located on the top of the hospital. The number of Aeromonas organisms isolated from tap water increased during the summer, with a peak in August, whereas few organisms were detected in the winter. The number of infections showed a similar seasonal pattern, and was inversely proportional to the number of hospital admissions. The mean monthly numbers of cases were 15 in summer and four in winter [26].

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Seasonality of Infections Caused by Gram-negative Bacilli
  7. HCAIs
  8. Discussion
  9. Transparency Declaration
  10. References

Many uncertainties exist regarding the extent of seasonality within the spectrum of infectious diseases and the mechanisms leading to seasonal variations. Therefore, the purpose of this review was to assess the existence of seasonal variations among infections caused by Gram-negative bacilli and nosocomial/healthcare-associated infections.

Regarding Gram-negative bacilli, seasonal variations have been demonstrated for infections, mostly BSIs, caused by Acinetobacter spp., E. coli, Enterobacter cloacae, Klebsiella spp., and P. aeruginosa, with higher rates of infections during the summer months in North America, Europe, the Middle East, Australia, and Asia [4–10]. In addition, a correlation has been observed between temperature increase and rates of BSI for Acinetobacter spp., P. aeruginosa, E. coli, K. pneumonia, and S. aureus. However, discrepancies exist in the results of studies on Klebsiella spp. BSIs. For Anderson et al. [8], K. pneumonia BSIs were subject to seasonal variations and correlated with climatic factors, whereas Al-Hasan et al. [9] found neither seasonal variation in BSI rates nor any association between those rates and average temperature. However, important differences exist between the designs of the two studies. Anderson’s study was hospital-based, whereas Al-Hasan’s was population-based, and it is possible that the choice of different denominators may have a profound impact on the detection of seasonal variations. Other differences existed. In Anderson’s study, only K. pneumoniae BSIs were included, whereas Al-Hasan also included K. oxytoca BSIs. Differences in temperature between Minnesota and the four geographical locations in Anderson’s study, where temperatures are significantly higher, may also have contributed to the difference in results (Table 1).

Table 1.  Seasonality in bloodstream infection and correlation with temperature
MicroorganismAuthor [reference]SeasonalityCorrelation with temperature
Acinetobacter spp.Eber [4]YesYes
Escherichia coli Eber [4]YesYes
Al Hassan [6]YesYes
Chazan [7]YesYes
Klebsiella pneumoniae Eber [4]YesYes
Anderson [8]YesYes
Al Hassan [9]NoNo
Enterobacter spp.Al Hassan [10]NoNo
Pseudomonas aeruginosa Eber [4]YesYes

Although the worldwide existence of seasonal variations in the rates of BSI for many bacterial species is striking, the studies performed provide more questions than answers.

Because BSI represents a complication of a focal infection, it would be useful to determine whether the climatic factors influence the occurrence of the primary infection (e.g. UTI) or influence the occurrence of complications within the population of infected patients. The study by Falagas et al. showing a significant correlation between medical visits for lower UTIs and average higher weekly temperature and decreased relative humidity could contribute to explain some of the seasonal variations observed with BSI, because, in many studies, UTI was the most common primary source of infection, e.g. up to 80% in Al-Hasan’s study on E. coli BSIs [6,11]. Another concern regarding these studies is the heterogeneity of the populations studied and the mixture of community-acquired and hospital-acquired infections. The proportion of patients with HCAI is not always presented. When it was known, the proportion of patients with community-acquired infections ranged from 59% in Al-Hasan’s study to 72% in Chazan’s study [6,7]. It is logical to suspect that factors influencing seasonal variations are not the same inside and outside the hospital. For instance, the nurse to patient ratio, which is often lower in summer, could play a role on the seasonality of hospital-acquired BSIs [27].

Overall, the most commonly described HCAIs that are subject to seasonal variations are viral and Acinetobacter spp. infections [12,16–19]. Nosocomial respiratory and gastrointestinal viral infection outbreaks are seasonal by nature, and noroviruses alone are responsible for >50% of all reported outbreaks of HCAI [12]. These outbreaks are associated with very high costs caused by the closure of affected units and staff absenteeism [14]. The other microorganisms causing HCAIs that are subject to seasonal variations are P. aeruginosa, E. coli, K. pneumoniae, Enterobacter cloacae, and Saureus [4,6–8,10,11].

All six studies on the epidemiology of Acinetobacter spp. have shown seasonal variations, and often the greatest seasonal variations of all bacterial species. Other interesting features of these infections are the variation in the increase in infection rates across the USA, with highest variations in the New England region, and the demonstration that seasonal increased incidence appears not to be caused by clonal dissemination of a single strain, but by multiple strains [4,15–19]. The reasons for the existence of seasonality in Acinetobacter. spp. HCAIs are unknown, but it is puzzling that it has been observed in US hospitals, where air-conditioning systems are widely used and no variations in temperature and humidity are supposed to occur (Table 2).

Table 2.  Seasonality in healthcare-associated infections and correlation with temperature
MicroorganismAuthor [reference]SeasonalityCorrelation with temperature
Enterobacter cloacae Perencevich [15]YesYes
Escherichia coli Perencevich [15]YesYes
Acinetobacter baumannii Perencevich [15]YesYes
Pseudomonas aeruginosa Perencevich [15]YesYes
Acinetobacter spp.Retailliau [16]Yes
McDonald [17]Yes
Gales [18]Yes
Christie [19]Yes

The other findings in HCAIs were the existence of summer peaks of surgical wound infection (SWI) in Finland and a positive correlation between carriage of ESBL-producing Enterobacteriaceae and mean temperature [22]. The latter result is intriguing, and prospective studies are needed.

The studies on CLABSI and peritoneal dialysis-associated peritonitis showed that those infections occurring in patients receiving care in an outpatient setting that are not strictly speaking ‘nosocomial’ are also subject to seasonal variations by different mechanisms. For CLABSI in Smith’s study, the BSI derived from contamination of the central catheter during recreational water exposure explained the occurrence of summer peaks of CLABSI in paediatric-oncology outpatients [20]. For peritonitis in patients with peritoneal dialysis, the physiopathology of the infection is less clear-cut, and various routes of infection may exist, including intraluminal, periluminal or haematogenous routes, and bacteria migrating through the bowel wall. In this instance also, further studies are needed [24,25].

The last article quoted is not the least important, as it demonstrates that the hospital water network may serve as a reservoir and source of bacteria that will proliferate in the summer and cause seasonal HCAI, such as the Aeromonas hydrophila responsible for HCAI in Picard’s study [26].

A limitation in many studies assessing seasonal variations resides in the way that data are analysed. In most studies, with the exception of Retailliau’s study on Acinetobacter spp., Kaier’s study on ESBL-producing Enterobacteriaceae, Kim’s study on peritoneal dialysis peritonitis, and Picard’s study on water-related Aeromonas hydrophila, infections were aggregated according to predefined seasons [16,21,22,24,26]. The problem with such an approach is that the aggregation of cases may result in a loss of information or hide periodicities in disease occurrences that are not strictly seasonal. In addition, the statistical tests used (chi-square tests or ANOVAs) are valid for independent events, but not for dependent events such as temperature variations or trends in infectious diseases [1]. Fisman, in his review on the seasonality of infectious diseases, and Zeger et al., in their review on time-series analysis of public health and biomedical data, suggest the use of temporal autocorrelation and time-series analysis to assess the seasonality of diseases [1,28]. Autocorrelation analysis was never performed, and time-series analyses were used in only three studies: the study by Perencevich et al. on summer peaks in the incidence rates of Gram-negative bacterial infections in hospitalized patients; the study by Eber et al. on Gram-negative BSIs; and the study by Kaier et al. on ESBL-producing Enterobacteriaceae [4,15,22].

In summary, seasonal variations have been shown for infections caused by many Gram-negative bacilli, as well as for a few HCAIs, but many studies remain to be performed in order to better understand the mechanisms of these variations.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Seasonality of Infections Caused by Gram-negative Bacilli
  7. HCAIs
  8. Discussion
  9. Transparency Declaration
  10. References
  • 1
    Fisman DN . Seasonality of infectious diseases . Annu Rev Public Health 2007 ; 28 : 127143 .
  • 2
    Dowell SF . Seasonal variation in host susceptibility and cycles of certain infectious diseases . Emerg Infect Dis 2001 ; 7 : 369374 .
  • 3
    Bronlee J . An investigation into the periodicity of measles epidemics in London from 1703 to the present day by the method of the periodogram . Phil Trans R Soc Lond 1918 ; 208 : 225250 .
  • 4
    Eber ME , Shardell M , Schweizer ML , Laxminarayan R , Perencevich EN . Seasonal and temperature-associated increases in Gram-negative bacterial bloodstream infections among hospitalized patients . PLoS ONE 2011 ; 6 : e25298 .
  • 5
    Gonçalves-Pereira J , Povoa PR , Lobo C , Carneiro AH . Bloodstream infections as a marker of community-acquired sepsis severity. Results from the Portuguese community-acquired sepsis study (SACiUCI study) . Clin Microbiol Infect 2012 ; doi: 10.1111/j.1469-0691.2012.03776.x. Jan 23. Epub ahead of print .
  • 6
    Al-Hassam MN , Lahr BD , Eckel-Passow JE , Baddour LM . Seasonal variation in Escherichia coli bloodstream infections: a population-based study . Clin Microbiol Infect 2009 ; 15 : 947950 .
  • 7
    Chazan B , Colodner R , Edelstein H , Raz R . Seasonal variations in Escherichia coli bloodstream infections in northern Israel . Clin Microbiol Infect 2011 ; 17 : 851854 .
  • 8
    Anderson DJ , Richet H , Chen LF et al. Seasonal variation in Klebsiella pneumoniae bloodstream infection on 4 continents . J Infect Dis 2008 ; 197 : 752756 .
  • 9
    Al-Hasan MN , Lahr BD , Eckel-Passow JE , Baddour LM . Epidemiology and outcome of Klebsiella species bloodstream infection: a population-based study . Mayo Clin Proc 2010 ; 85 : 139144 .
  • 10
    Al-Hasan MN , Lahr BD , Eckel-Passow JE , Baddour LM . Temporal trends in Enterobacter species bloodstream infection: a population-based study from 1998 to 2007 . Clin Microbiol Infect 2011 ; 17 : 539545 .
  • 11
    Falagas ME , Peppas G , Matthaiou DK , Karageorgopoulos DE , Karalis N , Theocharis G . Effect of meteorological variables on the incidence of lower urinary tract infections . Eur J Clin Microbiol Infect Dis 2009 ; 28 : 709712 .
  • 12
    Centers for Disease Control and Prevention . Surveillance for foodborne disease outbreaks—United States, 2006 . MMWR 2009 ; 58 : 609615 .
  • 13
    Spackova M , Altmann D , Eckmanns T , Koch J , Krause G . High level of gastrointestinal nosocomial infections in the German surveillance system, 2002–2008 . Infect Control Hosp Epidemiol 2010 ; 31 : 12731278 .
  • 14
    Danial J , Cepeda JA , Cameron F , Cloy K , Wishart D , Templeton KE . Epidemiology and costs associated with norovirus outbreaks in NHS Lothian, Scotland 2007–2009 . J Hosp Infect 2011 ; 79 : 354358 .
  • 15
    Perencevich EN , McGregor JC , Shardell M et al. Summer peaks in the incidences of Gram-negative bacterial infection among hospitalized patients . Infect Control Hosp Epidemiol 2008 ; 29 : 11241131 .
  • 16
    Retailliau HF , Hightower AW , Dixon RE , Allen JR . Acinetobacter calcoaceticus: a nosocomial pathogen with an unusual seasonal pattern . J Infect Dis 1979 ; 139 : 371375 .
  • 17
    McDonald LC , Banerjee SN , Jarvis WR , and The National Nosocomial Infections Surveillance System . Seasonal variation of Acinetobacter infections: 1987–1996 . Clin Infect Dis 1999 ; 29 : 11331137 .
  • 18
    Gales AC , Jones RN , Ferward KR , Linares J , Sadec HS , Verhoef J . Emerging importance of multidrug-resistant Acinetobacter species and Stenotrophomonas maltophilia as pathogens in seriously ill patients: geographic patterns, epidemiological features, and trends in the SENTRY Antimicrobial Surveillance Program (1997–1999) . Clin Infect Dis 2001 ; 32 ( suppl 21 ): 104113 .
  • 19
    Christie C , Mazon D , Hierholzer W Jr , Patterson JE . Molecular heterogeneity of Acinetobacter baumannii isolates during seasonal increase in prevalence . Infect Control Hosp Epidemiol 1995 ; 16 : 590594 .
  • 20
    Smith TL , Pullen GT , Crouse V , Rosenberg J , Jarvis WR . Bloodstream infections in pediatric oncology outpatients: a new healthcare systems challenge . Infect Control Hosp Epidemiol 2002 ; 23 : 239243 .
  • 21
    Koljonen V , Tukiainen E , Pipping D , Kolho E . Surgical site infections at Töölö hospital and the dog days myth . Duodecim 2009 ; 125 : 14151420 .
  • 22
    Kaier K , Frank U , Conrad A , Meyer E . Seasonal and ascending trends in the incidence of carriage of extended-spectrum β-lactamase-producing Escherichia coli and Klebsiella species in 2 German hospitals . Infect Control Hosp Epidemiol 2010 ; 31 : 11541159 .
  • 23
    Grassmann A , Gioberge S , Moeller S , Brown G . ESRD patients in 2004: global overview of patient numbers, treatment modalities and associated trends . Nephrol Dial Transplant 2005 ; 20 : 25872593 .
  • 24
    Kim MJ , Song JH , Park YJ , Kim GA , Lee SW . The influence of seasonal factors on the incidence of peritonitis in continuous ambulatory peritoneal dialysis in the temperate zone . Adv Perit Dial 2000 ; 16 : 243247 .
  • 25
    Cho Y , Badve SV , Hawley CM et al. Seasonal variation in peritoneal dialysis-associated peritonitis: a multi-centre registry study . Nephrol Dial Transplant 2011 ; 4 : 79 .
  • 26
    Picard B , Goullet P . Seasonal prevalence of nosocomial Aeromonas hydrophila infection related to Aeromonas in hospital water . J Hosp Infect 1987 ; 10 : 152155 .
  • 27
    Hugonnet S , Villaceves A , Pittet D . Nurse staffing level and nosocomial infections: empirical evaluation of the case-crossover and case-time-control designer . Am J Epidemiol 2007 ; 165 : 13211327 .
  • 28
    Zeger SL , Irizarry R , Peng RD . On time series analysis of public health and biomedical data . Annu Rev Public Health 2006 ; 27 : 5779 .