Study participants
During the study period there were 192,547 admissions to RCWMCH and 104 PABSI episodes. Analysis of the patient list and microbiology results obtained from the CDW and the NHLS clinical microbiology laboratory, GSH, respectively showed that during the study period 104 PABSI episodes occurred at RCWMCH. These episodes were used to estimate the risk of PABSI. Clinical and antibiotic data from 91 (87.5%) children were used in the subsequent analyses (Fig. 1).
Classification and risk of Pseudomonas aeruginosa BSI
A total of 104 PABSI episodes was identified among 103 study participants during the study period; 69 (66.3%) episodes were HAIs and 35 (33.7%) were IPOA. There was one recurrent PABSI episode, a HAI which manifested 30 days after the initial PABSI episode. The overall incidence risk of PABSI throughout the study period was 5.4 PABSI episodes / 10,000 hospital admissions. There was a decline in annual incidence risk from 2009 until 2016 followed by a rise in 2017. This increase was mainly related to an increase in the incidence risk of HAI (Fig. 2). The annual incidence risk of HAI was consistently higher than IPOA throughout the study period.
Characteristics of study participants
Table 1 describes characteristics of 91 PABSI episodes in 91 study participants, 60 (65.9%) episodes were HAIs and 31 (34.1%) were IPOA. The median time between admission and the development of PABSI in the 60 children who developed HAI was 13.5 days (IQR 7.0–28.0). The median age was 12 months (IQR 5–59) and 52% of the episodes occurred in females. A higher proportion of children with HAI (38/60, 63.3%), experienced antibiotic exposure during the 12-month period preceding PABSI compared to children with IPOA (9/31, 29.0%, p = 0.004). There were 13 cases of confirmed HIV infection with PABSI, 3/60 (5.0%) were HAI and 10/31 (32.3%) were IPOA. Fifteen (16.5%) of the children had underlying chronic illnesses other than HIV infection, that was present for at least 6 weeks duration. Eleven children had malignancies, namely, acute lymphoblastic leukaemia (3), acute myeloid leukaemia (2), lymphoma (1), neuroblastoma (1), craniopharyngioma (1) and germ cell tumour (1). Other chronic diseases included chronic lung disease (2), Fanconi anaemia (1) and aplastic anaemia (1). Children with IPOA had a significantly higher frequency of chronic diseases other than HIV, compared to those with HAI, (9/31, 29.0%) and (6/60, 10.0%) respectively, p = 0.034.
Table 1
Characteristics of children at the time of Pseudomonas aeruginosa bloodstream infection
Variable | HAI N = 60 | IPOA N = 31 | Total N = 91 | P value* |
Age (months) median (IQR) | 10 (4.5–46) | 19 (6–69) | 12 (5–59) | 0.065 |
Age category, n/N (%) | | | | |
<1 year | 34 (56.7) | 11 (35.5) | 45 (49.5) | 0.107 |
1–5 years | 11 (18.3) | 11 (35.5) | 22 (24.2) | |
>5 years | 15 (25.0) | 9 (29.0) | 24 (26.3) | |
Gender, n/N (%) | | | | |
Male | 27 (45.0) | 17 (55.8) | 44 (48.4) | 0.387 |
Female | 33 (55.0) | 14 (45.2) | 47 (51.6) | |
HIV status, n/N (%) | | | | |
HIV-infected | 3 (5.0) | 10 (32.3) | 13 (14.3) | 0.002 |
HIV-uninfected | 36 (60.0) | 15 (48.4) | 51 (56.0) | |
Unknown | 21 (35.0) | 6 (19.4) | 27 (29.7) | |
Weight-for-age, Z score category, n/N (%) | | | | |
Moderate underweight | 3 (5.2) | 6 (20.0) | 9 (10.2) | 0.008 |
Severe underweight | 11 (19.0) | 11 (36.7) | 22 (25.0) | |
Temperature in degrees Celsius | | | | |
<35⁰C | 1 (1.8) | 2 (6.7) | 3 (3.4) | 0.459 |
>35.5–37.9⁰C | 12 (21.1) | 7 (23.3) | 19 (21.8) | |
≥38.0⁰C | 44 (77.2) | 21 (70.0)) | 65 (74.7) | |
Anaemia | 40 (66.7) | 24 (77.4) | 64 (70.3) | 0.210 |
Hospitalisation in the 28-day period preceding the current admission, n/N (%) | 27 (45.0) | 20 (64.5) | 47 (51.6) | 0.121 |
Exposure to selective intravenous antibiotics, in the preceding 12 months, n/N (%) | 38 (63.3) | 9 (29.0) | 47 (51.7) | 0.004 |
Selective intravenous antibiotics exposure in the preceding 12 months, n/N (%) | | | | |
Gentamicin or amikacin | 19 (31.7) | 4 (12.4) | 23 (25.3) | 0.074 |
2nd to 4th generation cephalosporins | 8 (61.5) | 5 (38.5) | 13 (14.3) | 0.757 |
Piperacillin-tazobactam | 12 (20.0) | 0 (0) | 12 (13.2) | 0.005 |
Meropenem or ertapenem | 15 (25.0) | 1 (3.2) | 16 (17.6) | 0.009 |
Chronic diseases other than HIV infection, n/N (%) | 6 (10.0) | 9 (29.0) | 15 (16.5) | 0.034 |
ICU admission prior to PABSI, n/N (%) | 50 (83.3) | 11 (35.3) | 61 (67.0) | 0.0001 |
Central venous access device in situ, n/N (%) | 48 (80.0) | 11 (35.5) | 59 (54.8) | 0.0001 |
Endotracheal Intubation in situ, n/N (%) | 46 (76.7) | 8 (25.8) | 54 (59.3) | 0.0001 |
Burn wound, n/N (%) | 11 (18.3) | 8 (26.7) | 19 (21.1) | 0.416 |
Surgery during current admission, n/N (%) | 39 (65.0) | 7 (23.3) | 46 (51.1) | 0.0001 |
*Comparison of HAI and IPOA, HAI, healthcare-associated infection; IPOA, infections present on admission; PABSI, Pseudomonas aeruginosa bloodstream infection; C, Celsius; ICU, intensive care unit |
Presenting features and complications of PABSI
Respiratory distress was the commonest presenting feature overall (34/91, 37.4%) and for HAIs (23/60, 38.3%) whereas diarrhoea was the commonest presenting feature for IPOA (13/31, 41.9%). Shock as a presenting feature was significantly more frequent among children with IPOA (9/31, 29.0%) than HAI (4/60, 6.7%), p = 0.009. Pneumonia was the most frequent site of infection, occurring in 33/91 (36.3%) of the PABSI episodes (Table 2).
Table 2
Presenting features of Pseudomonas aeruginosa bloodstream infection and site of infection
Variable | HAI N = 60 n/N (%) | IPOA N = 31 n/N (%) | Total N = 91 n/N (%) | P value |
Presenting featuresa | | | | |
Respiratory distress | 23 (38.3) | 11 (35.5) | 34 (37.4) | 0.487 |
Diarrhoea | 16 (26.7) | 13 (41.9) | 29 (31.9) | 0.107 |
Wound infection | 18 (30.0) | 5 (16.1) | 23 (25.3) | 0.116 |
Shock | 4 (6.7) | 9 (29.0) | 13 (14.3) | 0.009 |
Ecythema gangrenosum | 1 (1.7) | 2 (6.5) | 3 (3.3) | 0.267 |
Otitis media | 1 (1.7) | 3 (9.7) | 4 (4.4) | 0.113 |
Otherb | 3 (5.0) | 4 (12.9) | 7 (7.7) | 0.224 |
Site of infection | | | | |
No definable focus | 12 (20.0) | 6 (19.4) | 18 (19.8) | 1.000 |
Pneumonia | 22 (36.7) | 11 (35.5) | 33 (36.3) | 1.000 |
Gastrointestinal tractc | 3 (5.0) | 4 (12.9) | 7 (7.7) | 0.224 |
Skin & soft tissue infection | 16 (26.7) | 4 (12.9) | 20 (22.0) | 0.184 |
Line infection | 6 (10.0) | 2 (6.5) | 8 (8.8) | 0.711 |
Urosepsis | 1 (1.7) | 3 (9.7) | 4 (4.4) | 0.132 |
HAI, healthcare-associated infection; IPOA, infection present on admission; a, some patients presented with > 1 presenting feature; b, vomiting (2), renal angle tenderness (1), necrotising bowel (1), tachycardia (1), eye discharge (1), acute abdomen (1); c, gastroenteritis (4), acute appendicitis (1), acute peritonitis (2).
Shock, as determined by the attending clinicians, was the commonest complication, occurring in 16/91 (17.6%) of the children; 11/16 (68.8%) required inotropic infusions. Shock, liver dysfunction and 14-day mortality were significantly more frequent complications in children with IPOA (Table 3).
Table 3
Complications and outcome associated with Pseudomonas aeruginosa bloodstream infection
Variable | HAI N = 60 n/N (%) | IPOA N = 31 n/N (%) | Total N = 91 n/N (%) | P value |
Shock | 7 (11.7) | 9 (29.0) | 16 (17.6) | 0.047 |
Coagulopathy | 5 (8.3) | 5 (16.1) | 10 (11.1) | 0.300 |
Renal dysfunction | 7 (11.7) | 5 (16.1) | 12 (13.2) | 0.534 |
Liver dysfunction | 0 (0) | 4 (12.9) | 4 (4.4) | 0.012 |
Respiratory failure | 6 (10.0) | 5 (16.1) | 11 (12.1) | 0.500 |
14-day mortality | 6 (10.0) | 11 (35.5) | 17 (18.7) | 0.005 |
HAI, healthcare-associated infection; IPOA, infection present on admission
Susceptibility profile of PA isolates and antibiotic therapy
Figure 3 summarises the susceptibility profile of the PA isolates during the study period. Overall, 69/91 (75.8%) of the PA isolates were susceptible to all antipseudomonal antibiotic categories evaluated, 12/91 (13.2%) isolates were MDR, 8/91 (8.8%) isolates were XDR and 2 (2.2%) isolates were PDR. The proportion of HAI isolates amongst the resistant isolates was 8/12 (66.7%) MDR, 7/8 (87.5%) XDR, and all the PDR isolates.
There were 19/91 (20.9%) isolates that were resistant to both imipenem and meropenem; there were an additional 2 that were resistant only to imipenem but not to meropenem i.e. a total of 21/91 (23.1%) isolates were resistant to imipenem. Ten meropenem-resistant isolates were susceptible to ceftazidime, (10/91; 9.1%), while 4/91 (4.4%) ceftazidime-resistant isolates were susceptible to meropenem. There were 8/91 (8.8%) isolates that were resistant to both ceftazidime and meropenem. Furthermore, of the 20/91 (22.0%) isolates resistant to gentamicin, 9/20 (45.0%) were susceptible to amikacin.
Fifty (54.9%) of the PABSI episodes were treated with appropriate empiric antibiotic therapy. A higher proportion of HAI PABSI episodes received appropriate empiric antibiotic therapy compared to IPOA PABSI episodes; 37/60 (61.7%) versus 13/31 (41.9%). This difference was, however, non-significant, p = 0.081. Three antibiotics frequently used in empiric therapy for both HAI and IPOA were meropenem 27/91(29.7%), piperacillin-tazobactam 19/91(20.9%), amikacin 18/91 (19.8%). Piperacillin-tazobactam was frequently combined with amikacin for empiric therapy 18/19 (94.7%). There were 7/31 (22.6%) IPOA episodes that were treated with empiric cephalosporins therapy; 5/7 (71.4%) received ceftriaxone and 2/7 (28.6%) cefotaxime. Six (19.4%) of the IPOA episodes received ampicillin monotherapy as empiric antibiotic therapy and in a further 5/31 (16.1%) episodes, ampicillin in combination with gentamicin, to which none of the isolates were susceptible. There were 3/31 (9.7%) IPOA episodes that were treated with empiric piperacillin-tazobactam plus amikacin, but the isolates were resistant to both antibiotics. In 23/60 (38.3%) episodes of HAI that received empiric antibiotic to which PA isolate was not susceptible to, 8/23 (34.8%) received ertapenem which is not an antipseudomonal antibiotic, 9/23 (39.1%) received piperacillin-tazobactam and amikacin and 6/23 (26.1%) received meropenem, to which the isolates were resistant.
The mean time ± SD to effective antibiotic therapy (appropriate empiric antibiotic therapy or definitive antibiotic therapy) was 1.3 days ± 1.1. The difference in mean time to effective antibiotic therapy for HAI was 1.2 days ± 0.9 and 1.6 days ± 1.5 in IPOA, this was not significant (p = 0.112). The antibiotic most frequently used as definitive antibiotic therapy for both IPOA and HAI was meropenem. Overall, 37/91 (40.7%) PABSI episodes were treated with meropenem; 23 of these 37 episodes (62.2%) isolates were caused by isolates that were susceptible to ceftazidime. By contrast, only 17.6% of the PABSI episodes were treated with ceftazidime (Table 4).
Table 4
Antibiotic susceptibility of Pseudomonas aeruginosa bloodstream infection isolates, and the definitive antibiotic therapy used during the study period
Antibiotic | Susceptibility of PA isolates to anti-pseudomonal antibiotics | Antibiotics used as definitive antibiotic therapy |
HAI N = 60 n/N (%) | IPOA N = 31 n/N (%) | Total N = 91 n/N (%) | HAI N = 60 n/N (%) | IPOA N = 31 n/N (%) | Total N = 91 n/N (%) |
Gentamicin | 43 (71.7) | 28 (90.3) | 71 (78.0) | 5 (8.3) | 2 (6.5) | 7 (7.7) |
Amikacin | 50 (83.3) | 28 (90.3) | 78 (85.7) | 10 (16.7) | 6 (19.4) | 16 (17.6) |
Ciprofloxacin | 42 (70.0) | 28 (90.3) | 70 (76.9) | 12 (20.0) | 7 (22.6) | 19 (20.9) |
Piperacillin-tazobactam | 35 (58.3) | 19 (61.3) | 54 (59.3) | 12 (20.0) | 9 (29.0) | 21 (23.1) |
Ceftazidime | 50 (83.3) | 29 (93.5) | 79 (86.8) | 8 (13.3) | 8 (25.8) | 16 (17.6) |
Cefepime | 47 (78.3 | 26 (83.9) | 73 (80.2) | 11 (18.3) | 2 (6.5) | 13 (14.3) |
Meropenem | 44 (73.3) | 28 (90.3) | 72 (79.1) | 27 (45.0) | 10 (32.3) | 37 (40.7) |
Imipenem Colistin | 43 (71.7) - | 27 (87.1) - | 70 (76.9) - | 2 (3.3) 9 (15) | 0 (0) - | 2 (2.2) 9 (9.9) |
PA, Pseudomonas aeruginosa; HAI, healthcare-associated infection; IPOA, infection present on admission
Outcome
There were 69/91 (75.8%) PABSI episodes that were successfully treated, with the children being discharged from hospital after these episodes. Twenty-two (24.2%) of the children died during their hospitalisation. Most of the deaths, 17/22 (77.3%), occurred within 14 days of hospitalisation as a direct result of PABSI. The median time (IQR) to death of these 17 children was 1.4 (1.0–8.3) days and 11/17 (64.7%) of these deaths were due to IPOA. Of the 5 deaths that occurred after 14 days, the median time (IQR) to death was 22.5 (20.8–30.3) days and 3/5 (60.0%) had IPOA.
On multivariable analysis, empiric antibiotic therapy to which PA isolate was not susceptible to, IPOA, and not being admitted in the ICU at the time that PABSI was diagnosed were significantly associated with 14-day mortality. Multidrug-resistant, XDR or PDR isolates were not predictors of 14-day mortality (see Table 5).
Table 5
Predictors of 14-day mortality in children with Pseudomonas aeruginosa bloodstream infection
Variable | Unadjusted OR (95% confidence interval) | P-value | Adjusted OR (95% confidence interval) | P value |
Age category N = 91 | | | | |
< 1 year | 1 | | | |
≥ 1 year | 0.66 (0.23–1.93) | 0.45 | | |
Weight N = 88 | | | | |
Normal weight | 1 | | | |
Moderate or severe underweight | 0.44 (0.15–1.32) | 0.10 | 1.60 (0.330–7.803) | 0.557 |
Admission in a health care facility 28 days prior to current hospitalisation | 0.80 (0.27–2.4) | 0.79 | | |
Chronic disease (excluding HIV) N = 80 | 1.2 (0.24–6.13) | 0.827 | | |
HIV status | | | | |
HIV-uninfected | 1 | | | |
HIV-infected | 1.34 (0.32–5.68) | 0.70 | | |
HIV status unknown | 0.89 (0.22–3.62) | 0.87 | | |
Nature of infection | | | | |
IPOA | 1 | | | |
HAI | 0.202 (0.066–0.619) | 0.005 | 0.083 (0.01–0.60) | 0.013 |
MDR, XDR or PDR isolates | 1.05 (0.30–3.61) | 0.945 | | |
Appropriate empiric antibiotic | 0.19 (0.06–0.63) | 0.007 | 0.23 (0.07–0.82) | 0.023 |
Burn wounds | 2.24 (0.47–10.77) | 0.315 | | |
Septic shock | 3.29 (1.05–10.23) | 0.040 | 2.27 (0.41–12.55) | 0.346 |
Anaemia | 1.47 (0.43–4.99) | 0.540 | - | |
Surgery during current admission | 1.96 (0.64-6.0) | 0.235 | - | |
ICU management | | | | |
Patient in the ICU prior to PABSI diagnosis | 0.43 (0.15–1.25) | 0.101 | 0.04 (0.005–0.365) | 0.004 |
Patient requiring ICU care after PABSI diagnosis | 1.21 (0.399–3.670) | 0.737 | - | |
OR, odds ratio; HAI, healthcare-associated infection; IPOA, infection present on admission; MDR, multidrug-resistant; XDR, extensively drug-resistant; PDR, pan drug-resistant |