During the seven years of surveillance, 153 incident BSI isolates of Klebsiella species were identified. In two cases infections had two different incident Klebsiella species concomitantly isolated and these included one patient with K. oxytoca and K. pneumoniae and another with K. oxytoca and K. variicola co-infection. There were therefore 151 incident Klebsiella species BSI episodes among 139 regional residents for an overall annual incidence rate of 12.1 per 100,000 per year. The incidence of K. pneumoniae and K. oxytoca were 9.1 and 2.9 per 100,000 per year, respectively. The remaining two incident isolates were K. variicola. Eleven patients had a second and one patient had a third episode of Klebsiella species BSI during the surveillance period. Twenty-four cases (16%) were hospital-onset, 90 (60%) were healthcare associated, and 37 (25%) were community-associated.
The annual incidence varied during the study years as shown in Figure 1. The first five years of the study demonstrated moderate year-to-year variability ranging from 6.8 to 11.7 per 100,000 annually but there was then a marked increase during the fifth and six years of the study (Figure 1). Although both species increased in incidence in the latter years of the study this was predominantly related to K. pneumoniae BSI in the last study year.
Risk factors and predisposing conditions
The median patient age was 71.4 (IQR, 58.8-80.9) years and 88 (58%) were male. No cases were observed among those aged less than 10 years and the incidence increased with older age as shown in Figure 2. Males were at higher risk but this was not statistically significant (14.0 vs. 10.1 per 100,000; IRR, 1.4; 95% CI, 1.0-1.9; p=0.06).
The median Charlson co-morbidity score was 2 (interquartile range, 1-4). Twenty-eight (19%) patients had a Charlson score of zero, and sixty-two (41%) had scores of 1-2, thirty-two (21%) 3-4, and twenty-nine (19%) had five or more. A number of co-morbidities were examined as risks for development of a Klebsiella species BSI within the population and these are shown in Table 1.
Clinical foci and microbiology
The most common focus of infection was intra-abdominal or genitourinary as shown in Table 2. Although K. pneumoniae was much more likely to be of hospital-onset as compared with K. oxytoca BSI (23/113; 20% vs. 1/35; 3%; relative risk 7.1; 95% confidence interval 1.0-50.9; p=0.016), otherwise the demographics and clinical features were similar (Table 2). Although there was a high rate of resistance to ampicillin, most isolates overall were susceptible to ceftriaxone, ciprofloxacin, gentamicin, and co-trimoxazole (Table 2). As compared to K. pneumoniae, K. oxytoca had a significantly reduced rate of susceptibility to cefazolin (Table 2).
Hospital admission and outcome
One hundred and forty-one (93%) patients were admitted to hospital for a median hospital length stay of 8 (IQR, 4-17) days. The in-hospital and 30-day all cause case-fatality rates were 24/141 (17%) and 27/151 (18%), respectively. The 30-day all cause case-fatality rates were 20% (23/113) and 9% (3/35; p=0.1) for K. pneumoniae and K. oxytoca, respectively.