The known: People with kidney failure often receive life‐sustaining haemodialysis treatment via central venous catheters (CVCs). Infections as complications of the haemodialysis CVC are, however, a major concern.
The new: The incidence of haemodialysis CVC infection‐related hospitalisations of adults with kidney failure in Australia is substantial, and it is higher among those under 70 years of age. Forty of 159 CVC‐related bloodstream infections were in people with functional alternative arteriovenous access for haemodialysis.
The implications: The incidence of haemodialysis CVC infections needs to be reduced in Australia. Timely removal of CVCs could be one strategy for achieving this aim.
Haemodialysis, the most frequent type of kidney replacement therapy, depends on reliable vascular access.1 For 60–80% of patients, maintenance haemodialysis commences with a central venous catheter (CVC);2,3 20–50% of people receiving maintenance haemodialysis have CVCs,4 and CVCs are increasingly considered for long term access in people with frailty or older than 80 years.5 Haemodialysis can be more rapidly established with a CVC than with an arteriovenous fistulas or a graft, but CVCs are more susceptible to infection‐related complications.6
Infections are a problem for people with haemodialysis CVCs,7 but the burden of infection attributable to CVCs is unclear. The reported incidence of haemodialysis CVC‐related bloodstream infections ranges between 0.6 and 6.5 episodes per 1000 catheter days,8 and the incidence has probably declined over the past three decades.9 The risks of Staphylococcus aureus bacteraemia10 and haemodialysis catheter‐related bloodstream infections11 were higher for people under 65 years of age in some studies, but not others.12,13 The impact of these infections may also differ between centres and countries. In a case series in a single United States centre, 34% of people with haemodialysis catheter‐related bloodstream infections were admitted to hospital;14 in a Western Australian study, 92% of people required hospital admission, but hospitalisation outcomes were not assessed.15
We therefore examined the epidemiology of catheter‐related infections in a national cohort of adults with kidney failure with incident haemodialysis CVCs in Australia. We specifically investigated whether people under 70 years of age were more frequently hospitalised with haemodialysis catheter‐related infections, and whether the impact of such hospitalisations was greater in this age group.
Methods
We undertook a cohort study nested within a cluster‐randomised trial. We analysed prospectively collected data from the national Reducing the burden of dialysis catheter complications (REDUCCTION) study, linked with Australian and New Zealand Dialysis and Transplant Registry (ANZDATA) data and hospitalisations data from all Australian states and territories except Western Australia.
Using prospectively collected REDUCCTION study data, we established a cohort of adults (18 years or older) who received incident CVCs for commencing or continuing chronic haemodialysis in Australia during 20 December 2016 – 31 March 2020 (Supporting Information, part 1).16,17 We excluded people not included in the ANZDATA registry (eg, people who received transient haemodialysis for acute kidney injury), whose first trial haemodialysis catheter was inserted for plasma exchange or an undocumented reason, and people enrolled in Western Australia (fewer than 5% of overall cohort; linked hospitalisation data were not available). Included people were followed from the date of insertion of their first CVC until the removal of their last CVC, death, or 31 March 2020, whichever was earliest. Baseline patient, catheter, and service characteristics were based on REDUCCTION data at the date of the first CVC insertion during the trial (Supporting Information, part 2).
Data linkage
Data for people enrolled in the REDUCCTION study were linked with the ANZDATA registry and hospital admissions datasets from all Australian states and territories except Western Australia using standard identifiers (name, gender, date of birth, medical record number; Supporting Information, part 3). Any hospitalisations commencing from one year prior to the start of the REDUCCTION study period (20 December 2015) to one year after its end (31 March 2021) were identified in the linked data.
Hospital admissions
Same‐day admissions and admissions for overnight haemodialysis only were excluded. Temporally continuous episodes of care in which patients were transferred to another hospital or to a rehabilitation facility were treated as single hospitalisations, and the principal diagnostic code (International Classification of Diseases, tenth revision, Australian modification; ICD‐10‐AM) for the initial admission was retained for the entire hospitalisation (Supporting Information, part 4). Hospitalisations with any type of catheter‐related infection were defined as those with a relevant ICD‐10‐AM code for either the principal diagnosis or as a secondary code (Supporting Information, table 1). We calculated the proportion of all infection‐related hospital admissions, categorised by principal ICD‐10‐AM diagnostic code (Supporting Information, table 2),18,19,20 that were attributed to haemodialysis catheter‐related infections.
Hospitalisations were defined as being caused by community‐onset haemodialysis catheter‐related bloodstream infections if the principal diagnostic code was compatible with a vascular access device infection or sepsis/bacteraemia, and a haemodialysis catheter‐related bloodstream infection event, which was adjudicated and confirmed during the trial, was reported during the three days preceding or the two days following admission (Supporting Information, part 5). During the trial, hospitals were advised to report all possible haemodialysis catheter‐related bloodstream infections; events were confirmed by an independent panel that judged cases using modified Infectious Diseases Society of America (IDSA) criteria16 (Supporting Information, part 1).
Length of hospital stay, requirement for intensive care unit (ICU) admission, time in ICU, in‐hospital mortality, and metastatic spread of the infection to distal sites during the same admission were assessed. A literature review established that infective endocarditis, osteomyelitis, septic arthritis, spinal discitis, and epidural abscess are the most frequent metastatic infection types after haemodialysis catheter‐related bloodstream infections;21,22,23,24,25 they were defined by compatible secondary diagnostic codes during the hospital admission caused by a haemodialysis catheter‐related bloodstream infection (Supporting Information, table 3). Time from haemodialysis catheter‐related bloodstream infection to CVC removal and whether the CVC required replacement were assessed. Admissions with principal diagnostic codes not compatible with sepsis/bacteraemia or vascular access device‐related infection, and admissions in which confirmed haemodialysis catheter‐related bloodstream infections occurred more than three days before or two days after admission, were deemed to have not been caused by haemodialysis catheter‐related bloodstream infections alone, and were therefore excluded from analyses of hospitalisations attributed to these infections.
Statistical analysis
We summarise baseline characteristics and hospitalisation outcomes as numbers and proportions, means with standard deviations (SDs), or medians with interquartile ranges (IQRs). Patient age was categorised as under 55 years, 55–70 years, and over 70 years. The incidence of hospitalisations involving haemodialysis CVC infections was estimated by dividing their number by total follow‐up time. Hospital length of stay was calculated from the index admission date. Follow‐up time was calculated as the number of days from first trial CVC insertion until the final CVC was removed, death, or 31 March 2020, whichever was earliest, plus one day (as some people had their only CVC inserted and removed on the same day). Follow‐up time and outcomes for people who moved to an older age category during follow‐up were partitioned by age category.
To assess associations between patient age category and the incidence of hospitalisation with any haemodialysis CVC infection, adjusted incidence rate ratios (IRRs) with 95% confidence intervals (CIs) were estimated using mixed effects negative binomial regression models. The relationship between age at admission with causative organisms was assessed using multivariable multinomial logistic regression; we reported adjusted risk ratios with 95% CIs. The relationship between log‐transformed length of stay for each first hospitalisation with haemodialysis CVC‐related bloodstream infection and causative organisms was assessed using multivariable linear regression (details for analyses: Supporting Information, part 6). Statistical analyses were undertaken in Stata/BE 18.0.
Ethics approval
The Sydney Local Health District Human Research Ethics Committee approved the study (2019/ETH07707), as did all relevant jurisdictions (Supporting Information, part 7). Our study adhered to the declaration of Helsinki.
Results
Of the 6248 patients enrolled in the REDUCCTION trial with incident haemodialysis CVCs during 20 December 2016 – 31 March 2020, we excluded 295 people enrolled in Western Australia, 1949 people without chronic kidney failure, and 61 who received their first haemodialysis CVC for plasma exchange or without a listed reason (Box 1). The mean age of the 3943 included people was 60.4 (SD, 15.5) years, 1556 were women (39.5%), and 485 were Aboriginal or Torres Strait Islander people (12.3%). Of the 2519 people who commenced chronic haemodialysis for incident kidney failure with CVCs, the transition was precipitated by acute kidney injury in 836 cases (33.2%; 21.2% of all people receiving incident CVC) (Box 2). In total, 6580 haemodialysis CVCs were inserted during the study period; 5296 were tunnelled (80.5%) and 4712 were in the right internal jugular vein (71.6%). The median number of haemodialysis CVCs per patient was one (IQR, 1–2), the median cumulative duration of use was 200 days (IQR, 93–362 days). The proportions of patients under 55 years of age or aged 55–70 years who were women or Aboriginal and Torres Strait Islander people were larger than for those over 70 years of age, and the proportions who had histories of cardiovascular disease or cancer were smaller (Box 2).
Incidence of hospitalisation with any haemodialysis catheter‐related infection
A total of 10 341 hospitalisations (overnight or longer) and 407 deaths were recorded over 2633.7 patient‐years of follow‐up; multiple‐day hospitalisations were not identified for 168 patients (4.3%). Catheter‐related infections were coded for 644 hospitalisations (24.5 [95% CI, 22.6–26.4] per 100 patient‐years); the incidence was higher for people under 55 years of age (adjusted IRR, 1.55; 95% CI, 1.21–1.98) and for people aged 55–70 years (adjusted IRR, 1.34; 95% CI, 1.05–1.70) than for people over 70 years of age (Box 3). Sensitivity analyses in which hospitalisations with haemodialysis CVC‐related infection was defined by the T82.77 diagnostic code only yielded similar results (Supporting Information, table 6).
Hospitalisations caused by infections
Based on principal ICD‐10‐AM diagnostic codes, 1938 (18.7%) of 10 341 hospitalisations during 20 December 2016 – 31 March 2020 were primarily caused by infections (Box 4), most frequently vascular access device infection (456 hospitalisations), sepsis/bacteraemia (421), pneumonia (375), intra‐abdominal infections (297), cellulitis (138), and bone/joint infections (91 hospitalisations). For 110 of 456 hospitalisations attributed to vascular access device infection (24.1%) and 49 of 421 attributed to sepsis/bacteraemia (11.6%), trial‐confirmed haemodialysis catheter‐related bloodstream infections between three days before and two days after the initial admission were recorded; these 159 hospitalisations comprised 93.5% of the 170 community‐onset haemodialysis catheter‐related bloodstream infections reported and verified during the trial and 8.2% of 1938 infection‐related hospitalisations (Box 4); they accounted for 2181 of 18 913 infection‐related hospitalisations bed‐days (11.5%) (Box 5). Admissions caused by community‐onset haemodialysis catheter‐related bloodstream infections were responsible for 8.8% of all infection‐related hospitalisations of people under 55 years of age (11.7% of bed‐days), 9.7% of those of people aged 55–70 years (13.3% of bed‐days), and 6.2% of those of people over 70 years of age (9.4% of bed‐days) (Box 6).
Haemodialysis catheter‐related bloodstream infection events were also reported during 71 further hospitalisations (Box 4): 27 cases with other principal diagnostic codes, 16 cases in which the infection developed more than two days after admission, and 28 cases in which the infection developed during an admission that precipitated the initial CVC insertion. Eleven haemodialysis catheter‐related bloodstream infections were not associated with hospitalisations. The ICD‐10‐AM diagnostic code T82.77 was recorded for 92 of 159 hospitalisations caused by haemodialysis catheter‐related bloodstream infections (57.9%), 13 of 43 complicated by haemodialysis catheter‐related bloodstream infections (30%), and 254 of 10 139 without trial‐confirmed haemodialysis catheter‐related bloodstream infections (2.5%) (Supporting Information, table 8).
Hospitalisations caused by haemodialysis catheter‐related bloodstream infections
The 159 hospitalisations caused by haemodialysis catheter‐related bloodstream infections involved 151 patients; four were hospitalised twice, two were hospitalised three times. Staphylococcus aureus was the organism most frequently identified (77 hospitalisations). The median hospitalisation stay was ten (IQR, 5–15) days; metastatic spread of infection was detected in twelve cases (7.5%), including five infective endocarditis, four septic joint, and three osteomyelitis events; four people died in hospital. Nineteen of 121 hospitalisations for which the information was available (15.7%) included ICU admissions (median stay, 2.7 days; IQR, 1.1–4.6 days). The infected CVC was removed within seven days of infection during 140 of 159 admissions (88.1%); 40 removed CVCs (28.6%) were not replaced because a functioning arteriovenous fistula or graft was present (Supporting Information, table 9).
The mean age of 151 patients at the time of their first hospitalisation caused by haemodialysis catheter‐related bloodstream infections was 59.2 (SD, 15.5) years; 114 (75.5%) were under 70 years of age (Box 7). The risk of haemodialysis catheter‐related bloodstream Staphylococcus aureus infection hospitalisation declined with age (per decade: adjusted relative risk ratio, 0.65; 95% CI, 0.47–0.89), in contrast to haemodialysis catheter‐related bloodstream infections caused by other Staphylococcus spp. and Gram‐negative bacteria (Box 8; Supporting Information, table 10). The mean length of hospital stays caused by Staphylococcus aureus (13.5 [95% CI, 7.5–17] days) or Streptococcus spp. infections (22.5 [95% CI, 13.5–37] days) was longer than for those caused by other pathogens (other Staphylococcus spp.: 8 [95% CI, 4–14] days; Gram‐negative bacteria: 6 [95% CI, 4–9] days; Supporting Information, table 11).
Discussion
During 2016–20, the incidence of haemodialysis CVC infection‐related hospitalisations among adults with kidney failure in Australia was substantial, and was higher among those under 70 years of age. Of 170 people with community‐onset haemodialysis catheter‐related bloodstream infections, 159 were admitted to hospital (93.5%); these hospitalisations comprised 8.2% of all infection‐related hospitalisations. The median length of hospital stay was ten days (IQR, 5–15 days), metastatic spread of infection was detected in twelve cases (7.5%), and four people died in hospital (2.5%).
The baseline characteristics of people who started incident haemodialysis with CVCs were similar to those recorded in ANZDATA for all people in Australia receiving haemodialysis,2 and did not indicate clear contraindications for arteriovenous access. According to vascular access guidelines, age, frailty, and short life expectancy are important factors when considering long term CVCs,5 but in our study 69% of incident haemodialysis CVCs were inserted into people under 70 years of age. Interventions that facilitate earlier referral to nephrology services or improve the success rate of arteriovenous access surgery could reduce the need for CVC use, avert infection‐related complications, and improve vascular access outcomes. Genuine engagement and collaborative research with Aboriginal and Torres Strait Islander people are essential for assessing and reducing disparities in their outcomes.26
The higher incidence of haemodialysis CVC‐related infections in people under 70 years of age we found is consistent with other reports. People under 65 years of age and adults from ethnic minorities receiving maintenance haemodialysis are more likely than other people to experience S. aureus bacteraemia,10 which is associated with poorer outcomes; people over 75 years of age less frequently experience bloodstream infections attributable to their haemodialysis CVC.11 Although other studies have not found these relationships,12,13 it is plausible that the risk of hospitalisation with haemodialysis CVC‐related infections is higher for people under 70 years, as they may be more active and have more contact with others in the community, and their behaviour may place them at greater risk of infection than older people, resulting in more frequent catheter colonisation and infection. People under 60 years have a greater capacity for sweating,27 which could favour bacterial growth on moist skin. Further, nasal colonisation with S. aureus is more frequent in people under 60 years of age,28,29 who may therefore be more susceptible to haemodialysis catheter‐related bloodstream S. aureus infections.23 Interventions that prevent haemodialysis CVC infections are needed, particularly S. aureus infections in younger people.
The characteristics of hospitalisations with confirmed haemodialysis catheter‐related bloodstream infections in Australia differed from those reported in the United States, where fewer than 50% of people with haemodialysis catheter‐related bloodstream infections were hospitalised, and the median length of stay was 4–5 days.14,23,30 However, metastatic spread of infection, the need for ICU admission, and in‐hospital mortality were high in both countries. The frequency of haemodialysis catheter‐related bloodstream infections as complications of hospitalisations for other reasons has not previously been reported. As we found that they accounted for 71 of 241 trial‐reported haemodialysis catheter‐related bloodstream infections (29.4%), they should be investigated further. Research into factors involved in differences between countries, such as outpatient use of intravenous antibiotics and preferred choice and route of antimicrobial administration, is needed to assess the relative benefits and risks of different treatment strategies.
The earlier identification and removal of haemodialysis CVCs that are no longer required because alternative functional dialysis access is available could prevent as many as 25% of hospitalisations attributed to haemodialysis catheter‐related bloodstream infections. We found that 40 of 140 CVCs removed within seven days of identified infections did not require replacement because functioning arteriovenous access was available. The incidence of haemodialysis CVC infections in people with alternative dialysis access has not previously been assessed, but delays in removal could result in unnecessary bloodstream infections.31 Given natural variation in rates of wound healing and maturation after establishing arteriovenous access, the standard waiting time could be reduced for some patients. Haemodialysis CVC removal practices should be further investigated.
Limitations
First, we may have underestimated the incidence of haemodialysis CVC‐related infections. As linked hospitalisations were not identified for a small number of people with incident CVCs, we may have missed some hospitalisations with haemodialysis CVC‐related infections. The estimated incidence of hospitalisations attributable to haemodialysis catheter‐related bloodstream infections did not include infections that complicated hospitalisations with other causes, or clinically suspected haemodialysis catheter‐related bloodstream infections that did not satisfy the modified IDSA criteria. Second, the data for this study were collected during a national stepped wedge cluster randomised trial that sought to reduce the incidence of haemodialysis CVC‐related bloodstream infections in Australia. The trial intervention was not effective, but a modest decline in reported infections was found.9,16 The estimated burden of infection in our study is therefore a weighted mean of that during the entire trial period, and could be lower than if the trial had not been undertaken. Further studies are needed to better quantify the burden of haemodialysis CVC infections, and to assess whether practice changes, including proceduralist factors, influence the infection risk. Third, the numbers of metastatic infections, ICU admissions, and in‐hospital deaths among people hospitalised with haemodialysis catheter‐related bloodstream infections were too small to assess their frequency by age group. Finally, economic modelling is needed to assess the cost of haemodialysis CVC infections in Australia.22,30,32
Conclusion
The health burden of haemodialysis CVC infections in Australia is substantial, particularly among adults under 70 years of age with incident haemodialysis CVCs. Timely removal of unnecessary haemodialysis CVCs could reduce the number of infections, which would improve the lives of people with kidney failure, particularly those under 70 years of age.
Box 1 – Derivation of study cohort from the Reducing the burden of dialysis catheter complications (REDUCCTION) study cohort, and their transition to haemodialysis with central venous catheters
Box 2 – Baseline characteristics of 3943 people with incident haemodialysis central venous catheter, by age group*
|
Characteristic |
All people |
Under 55 years |
55–70 years |
Over 70 years |
|||||||||||
|
|
|||||||||||||||
|
Number |
3943 |
1335 |
1381 |
1227 |
|||||||||||
|
Age (years), mean (SD) |
60.4 (15.5) |
42.6 (9.8) |
62.9 (4.3) |
77.0 (4.9) |
|||||||||||
|
Gender (women) |
1556 (39.5%) |
574 (43.0%) |
556 (40.3%) |
426 (34.7%) |
|||||||||||
|
Ethnic background |
|
|
|
|
|||||||||||
|
White |
2378 (60.3%) |
712 (53.3%) |
816 (59.1%) |
850 (69.3%) |
|||||||||||
|
Aboriginal or Torres Strait Islander |
485 (12.3%) |
282 (21.1%) |
172 (12.5%) |
31 (2.5%) |
|||||||||||
|
Asian |
364 (9.2%) |
108 (8.1%) |
136 (9.8%) |
120 (9.8%) |
|||||||||||
|
Māori or Pasifika |
152 (3.9%) |
53 (4.0%) |
71 (5.1%) |
28 (2.3%) |
|||||||||||
|
Other |
564 (14.3%) |
180 (13.5%) |
186 (13.5%) |
198 (16.1%) |
|||||||||||
|
Private hospital insurance |
1029 (26.1%) |
254 (19.0%) |
374 (27.1%) |
401 (32.7%) |
|||||||||||
|
Never smoked |
1941 (49.2%) |
666 (49.9%) |
643 (46.6%) |
632 (51.5%) |
|||||||||||
|
Primary kidney disease |
|
|
|
|
|||||||||||
|
Diabetic kidney disease |
1550 (39.3%) |
468 (35.1%) |
638 (46.2%) |
444 (36.2%) |
|||||||||||
|
Glomerular disease |
730 (18.5%) |
364 (27.3%) |
191 (13.8%) |
175 (14.3%) |
|||||||||||
|
Hypertension |
422 (10.7%) |
86 (6.4%) |
120 (8.7%) |
216 (17.6%) |
|||||||||||
|
Polycystic kidney disease or reflux nephropathy |
272 (6.9%) |
122 (9.1%) |
98 (7.1%) |
52 (4.2%) |
|||||||||||
|
Other |
912 (23.1%) |
272 (20.4%) |
312 (22.6%) |
328 (26.7%) |
|||||||||||
|
Not reported |
57 (1.4%) |
23 (1.7%) |
22 (1.6%) |
12 (1.0%) |
|||||||||||
|
Immunosuppressant use |
513 (13.0%) |
228 (17.1%) |
176 (12.7%) |
109 (8.9%) |
|||||||||||
|
Medical conditions |
|
|
|
|
|||||||||||
|
Coronary artery disease |
1164 (29.5%) |
218 (16.3%) |
439 (31.8%) |
507 (41.3%) |
|||||||||||
|
Ever had cancer |
670 (17.0%) |
82 (6.1%) |
239 (17.3%) |
349 (28.4%) |
|||||||||||
|
Peripheral arterial disease |
621 (15.7%) |
152 (11.4%) |
243 (17.6%) |
226 (18.4%) |
|||||||||||
|
Chronic lung disease |
468 (11.9%) |
106 (7.9%) |
178 (12.9%) |
184 (15.0%) |
|||||||||||
|
Previous stroke |
345 (8.7%) |
65 (4.9%) |
136 (9.8%) |
144 (11.7%) |
|||||||||||
|
Indication for first central venous catheter |
|
|
|
|
|||||||||||
|
Start maintenance haemodialysis |
1683 (42.7%) |
628 (47.0%) |
588 (42.6%) |
467 (38.1%) |
|||||||||||
|
Acute kidney injury |
836 (21.2%) |
259 (19.4%) |
316 (22.9%) |
261 (21.3%) |
|||||||||||
|
Arteriovenous access complication |
763 (19.4%) |
226 (16.9%) |
258 (18.7%) |
279 (22.7%) |
|||||||||||
|
Transfer from peritoneal dialysis |
625 (15.9%) |
204 (15.3%) |
208 (15.1%) |
213 (17.4%) |
|||||||||||
|
Failed transplant |
36 (0.9%) |
18 (1.3%) |
11 (0.8%) |
7 (0.6%) |
|||||||||||
|
First catheter tunnelled |
3029 (76.8%) |
984 (73.7%) |
1063 (77.0%) |
982 (80.0%) |
|||||||||||
|
First catheter in right internal jugular vein |
3068 (77.8%) |
1049 (78.6%) |
1071 (77.6%) |
948 (77.3%) |
|||||||||||
|
Proceduralist responsible for insertion |
|
|
|
|
|||||||||||
|
Interventional radiology |
2074 (52.6%) |
636 (47.6%) |
730 (52.9%) |
708 (57.7%) |
|||||||||||
|
Critical care |
627 (15.9%) |
205 (15.4%) |
221 (16.0%) |
201 (16.4%) |
|||||||||||
|
Nephrologists |
618 (15.7%) |
292 (21.9%) |
226 (16.4%) |
100 (8.1%) |
|||||||||||
|
Surgeons |
543 (13.8%) |
174 (13.0%) |
175 (12.7%) |
194 (15.8%) |
|||||||||||
|
Other |
81 (2.1%) |
28 (2.1%) |
29 (2.1%) |
24 (2.0%) |
|||||||||||
|
State of enrolment |
|
|
|
|
|||||||||||
|
New South Wales/Australian Capital Territory |
1217 (30.9%) |
364 (27.3%) |
410 (29.7%) |
443 (36.1%) |
|||||||||||
|
Victoria or Tasmania |
1089 (27.6%) |
329 (24.6%) |
360 (26.1%) |
400 (32.6%) |
|||||||||||
|
Queensland |
943 (23.9%) |
318 (23.8%) |
364 (26.4%) |
261 (21.3%) |
|||||||||||
|
South Australia |
398 (10.1%) |
135 (10.1%) |
157 (11.4%) |
106 (8.6%) |
|||||||||||
|
Northern Territory |
296 (7.5%) |
189 (14.2%) |
90 (6.5%) |
17 (1.4%) |
|||||||||||
|
Regional enrolment site |
571 (14.5%) |
255 (19.1%) |
207 (15.0%) |
109 (8.9%) |
|||||||||||
|
Large service size† |
1677 (42.5%) |
680 (50.9%) |
585 (42.4%) |
412 (33.6%) |
|||||||||||
|
|
|||||||||||||||
|
SD = standard deviation. * At time of enrolment in the REDUCCTION study. Baseline characteristics for people with incident kidney failure (2519 people) or prevalent kidney failure (1424 people) are included in the Supporting Information, table 4; baseline characteristics for people with a central venous catheter that was not precipitated by an acute kidney injury, by age group, are included in the Supporting Information, table 5. † Large service size: more than 250 patients receiving haemodialysis on 31 December 2016. |
|||||||||||||||
Box 3 – Hospitalisations with any haemodialysis catheter‐related infection among 3943 people with incident haemodialysis central venous catheters, by age group*
|
Characteristic |
All people |
Under 55 years |
55–70 years |
Over 70 years |
|||||||||||
|
|
|||||||||||||||
|
Number of people |
3943 |
1335 |
1381 |
1227 |
|||||||||||
|
Total patient follow‐up time (days) |
961 974 |
302 177 |
322 099 |
337 698 |
|||||||||||
|
Hospitalisations with any haemodialysis central venous catheter infection |
|
|
|
|
|||||||||||
|
Number |
644 |
235 |
219 |
190 |
|||||||||||
|
Incidence, per 100 patient‐years (95% CI) |
24.5 (22.6–26.4) |
28.4 (24.8–32.0) |
24.8 (21.5–28.1) |
20.6 (17.6–23.5) |
|||||||||||
|
Adjusted incidence rate ratio (95% CI)† |
— |
1.55 (1.21–1.98) |
1.34 (1.05–1.70) |
1 |
|||||||||||
|
Deaths |
407 (10.3%) |
66 (4.9%) |
147 (10.6%) |
194 (15.8%) |
|||||||||||
|
|
|||||||||||||||
|
CI = confidence interval. * At time of enrolment in the REDUCCTION study. † Mixed effects negative binomial regression model, adjusted for gender, ethnic background, private hospital insurance, smoking history, indication for the first haemodialysis catheter, immunosuppressant use, primary kidney disease, and atherosclerotic cardiovascular disease. |
|||||||||||||||
Box 4 – All‐cause multiple‐day hospitalisations and confirmed haemodialysis catheter‐related bloodstream infections among 3943 people with incident haemodialysis central venous catheters*
* No linked multiple‐day hospitalisations were identified for 168 people in the study cohort.
Box 5 – Hospital admissions caused by infections in people with incident haemodialysis central venous catheters, and associated number of bed‐days*
* The data underlying this figure are included in the Supporting Information, table 7.
Box 6 – Proportions of infection‐related hospitalisations and hospital bed‐days attributable to community‐onset haemodialysis catheter‐related bloodstream infections, by age group
|
|
Reason for hospital admission |
||||||||||||||
|
Age group at admission |
Any infection |
Haemodialysis catheter‐related bloodstream infection |
|||||||||||||
|
|
|||||||||||||||
|
Hospital admissions |
1938 |
159 (8.2%) |
|||||||||||||
|
Under 55 years |
650 |
57 (8.8%) |
|||||||||||||
|
55–70 years |
640 |
62 (9.7%) |
|||||||||||||
|
Over 70 years |
648 |
40 (6.2%) |
|||||||||||||
|
Time in hospital (days) |
18 913 |
2181 (11.5%) |
|||||||||||||
|
Under 55 years |
5653 |
663 (11.7%) |
|||||||||||||
|
55–70 years |
7002 |
929 (13.3%) |
|||||||||||||
|
Over 70 years |
6258 |
589 (9.4%) |
|||||||||||||
|
|
|||||||||||||||
|
|
|||||||||||||||
Box 7 – Patient characteristics and outcomes of first hospitalisation caused by haemodialysis catheter‐related bloodstream infections, by age group*
|
Characteristic |
All people |
Under 55 years |
55–70 years |
Over 70 years |
|||||||||||
|
|
|||||||||||||||
|
Number |
151 |
54 |
60 |
37 |
|||||||||||
|
Age (years), mean (SD) |
59.2 (15.5) |
42.3 (9.3) |
63 (4.6) |
77.6 (5.6) |
|||||||||||
|
Gender (women) |
57 (37.7%) |
26 (48%) |
20 (33%) |
11 (30%) |
|||||||||||
|
Ethnic background |
|
|
|
|
|||||||||||
|
White |
103 (68.2%) |
34 (63%) |
43 (72%) |
26 (70%) |
|||||||||||
|
Aboriginal or Torres Strait Islander |
14 (9.3%) |
8 (15%) |
6 (10%) |
0 |
|||||||||||
|
Asian |
12 (7.9%) |
4 (7.4%) |
3 (5.0%) |
5 (14%) |
|||||||||||
|
Māori and Pasifika |
2 (1.3%) |
1 (1.9%) |
1 (1.7%) |
0 |
|||||||||||
|
Other |
20 (13.2%) |
7 (13%) |
7 (12%) |
6 (16%) |
|||||||||||
|
Private hospital insurance |
33 (21.9%) |
4 (7.4%) |
18 (30%) |
11 (30%) |
|||||||||||
|
Never smoked |
69 (45.7%) |
24 (44%) |
28 (47%) |
17 (46%) |
|||||||||||
|
Primary kidney disease |
|
|
|
|
|||||||||||
|
Diabetic kidney disease |
51 (33.8%) |
16 (30%) |
27 (45%) |
8 (22%) |
|||||||||||
|
Glomerular disease |
34 (22.5%) |
19 (35%) |
10 (17%) |
5 (14%) |
|||||||||||
|
Hypertension |
23 (15.2%) |
5 (9.3%) |
8 (13%) |
10 (27%) |
|||||||||||
|
Polycystic kidney disease or reflux nephropathy |
13 (8.6%) |
8 (15%) |
3 (5.0%) |
2 (5.4%) |
|||||||||||
|
Other |
21 (13.9%) |
6 (11%) |
9 (15%) |
6 (16%) |
|||||||||||
|
Not reported |
9 (6.0%) |
0 |
3 (5.0%) |
6 (16%) |
|||||||||||
|
Immunosuppressant use |
15 (9.9%) |
8 (15%) |
5 (8.3%) |
2 (5.4%) |
|||||||||||
|
Medical conditions |
|
|
|
|
|||||||||||
|
Ever had cancer |
19 (12.6%) |
1 (1.9%) |
7 (12%) |
11 (30%) |
|||||||||||
|
Coronary artery disease |
50 (33.1%) |
12 (22%) |
21 (35%) |
17 (46%) |
|||||||||||
|
Peripheral arterial disease |
27 (17.9%) |
8 (15%) |
10 (17%) |
9 (24%) |
|||||||||||
|
Previous stroke |
11 (7.3%) |
3 (5.6%) |
4 (6.7%) |
4 (11%) |
|||||||||||
|
Chronic lung disease |
15 (9.9%) |
4 (7.4%) |
7 (12%) |
4 (11%) |
|||||||||||
|
Microbiology |
|
|
|
|
|||||||||||
|
Staphylococcus aureus |
72 (47.7%) |
31 (57%) |
29 (48%) |
12 (32%) |
|||||||||||
|
Other Staphylococcus spp. (including S. epidermidis) |
18 (11.9%) |
5 (9.3%) |
5 (8.3%) |
8 (22%) |
|||||||||||
|
Streptococcus species (including Enterococcus spp.) |
12 (7.9%) |
4 (7.4%) |
5 (8.3%) |
3 (8.1%) |
|||||||||||
|
Gram‐negative bacteria† |
35 (23.2%) |
8 (15%) |
16 (27%) |
11 (30%) |
|||||||||||
|
Fungal |
4 (2.6%) |
1 (1.9%) |
3 (5.0%) |
0 |
|||||||||||
|
Polymicrobial |
10 (6.6%) |
5 (9.3%) |
2 (3.3%) |
3 (8.1%) |
|||||||||||
|
Time to catheter removal (days), median (IQR) |
2.5 (1.5–3.5) |
2 (1.5–3.5) |
2.5 (1.5–4.5) |
1.5 (1.5–3.5) |
|||||||||||
|
Catheter removed within seven days |
133 (88.1%) |
48 (89%) |
53 (88%) |
32 (86%) |
|||||||||||
|
Catheter removed but not replaced (functioning arteriovenous access) |
38 (25.2%) |
13 (24%) |
19 (32%) |
6 (16%) |
|||||||||||
|
Length of stay (days), median (IQR) |
10 (5–16) |
9 (4–15) |
11 (5.5–15) |
10 (6–18) |
|||||||||||
|
Metastatic complications |
|
|
|
|
|||||||||||
|
Infective endocarditis |
5 (3.3%) |
2 (3.7%) |
3 (5.0%) |
0 |
|||||||||||
|
Osteomyelitis |
3 (2.0%) |
0 |
2 (3.3%) |
1 (2.7%) |
|||||||||||
|
Septic joint |
2 (1.3%) |
0 |
1 (1.7%) |
1 (2.7%) |
|||||||||||
|
Admission to intensive care unit‡ |
19/109 (12.6%) |
8/43 (19%) |
6/43 (14%) |
5/23 (22%) |
|||||||||||
|
Intensive care unit stay (days), median (IQR) |
2.7 (1.1–4.6) |
0.83 (0.1–1.7) |
3.6 (1.4–4.6) |
3.3 (3.0–3.6) |
|||||||||||
|
Died in hospital |
4 (2.6%) |
1 (1.9%) |
3 (5.0%) |
0 |
|||||||||||
|
|
|||||||||||||||
|
IQR = interquartile range, SD = standard deviation. * Four people were hospitalised twice, and two three times with haemodialysis catheter‐related bloodstream infections; in each case, only their first hospitalisation is included here. † Escherichia coli, Enterobacter cloacae, Enterobacter aerogens, Klebsiella pneumoniae, Klebsiella oxytoca, Klebsiella variicola, Pseudomonas aeruginosa, Pseudomonas luteola, Serratia marcescens, Serratia liquefaciens, Serratia spp., Haemophilus influenzae, Acinetobacter baumanii complex, Citrobacter freundii, Achromobacter spp., Herbaspirillum huttiense. ‡ Information not available for all patients. |
|||||||||||||||
Received 30 September 2024, accepted 10 February 2025
- Benjamin Lazarus1,2,3,4
- Kevan R Polkinghorne2,4
- Martin P Gallagher1
- Jayson Catiwa1,5
- Nicholas A Gray6
- Sarah Coggan1
- Kathryn R Higgins1
- Girish Talaulikar7,8
- Stephen P McDonald AM9,10
- Sradha Kotwal1,5
- 1 The George Institute for Global Health, Sydney, NSW
- 2 Monash University, Melbourne, VIC
- 3 Centre for Health Services Research, the University of Queensland, Brisbane, QLD
- 4 Monash Medical Centre, Melbourne, VIC
- 5 Prince of Wales Hospital, Sydney, NSW
- 6 Sunshine Coast University Hospital, Birtinya, QLD
- 7 Canberra Hospital, Canberra, ACT
- 8 College of Medicine, Australian National University, Canberra, ACT
- 9 Australia and New Zealand Dialysis and Transplant (ANZDATA) Registry, Adelaide, SA
- 10 University of Adelaide, Adelaide, SA
Correspondence: skotwal@georgeinstitute.org.au
Open access:
Open access publishing facilitated by The University of Queensland, as part of the Wiley – The University of Queensland agreement via the Council of Australian University Librarians.
Data Sharing:
The individual patient data generated in the trial can be shared in accordance with the trial's data sharing policy and in accordance with the local regulatory and ethics approval for the trial.
We thank all investigators, nephrology services, and patients for their participation in the trial. The study was funded by the National Health and Medical Research Council (NHMRC; APP1103241, APP2005174, MRF1150335). The NHMRC was not involved in study design, data collection, analysis or interpretation, reporting, or publication.
No relevant disclosures.
- 1. Bello AK, Okpechi IG, Osman MA, et al. Epidemiology of haemodialysis outcomes. Nat Rev Nephrol 2022; 18: 378‐395.
- 2. Australia and New Zealand Dialysis and Transplant Registry. Haemodialysis. In: ANZDATA annual report 2022. https://www.anzdata.org.au/wp‐content/uploads/2023/05/Chapter‐4‐Haemodialysis‐ANZDATA‐Annual‐Report‐2022.pdf (viewed Sept 2023).
- 3. Pisoni RL, Zepel L, Port FK, Robinson BM. Trends in US vascular access use, patient preferences, and related practices: an update from the US DOPPS Practice Monitor with international comparisons. Am J Kidney Dis 2015; 65: 905‐915.
- 4. Lazarus B, Lok CE, Moist L, Polkinghorne KR. Strategies to prevent hemodialysis catheter dysfunction. J Am Soc Nephrol 2025; 36: 952‐966.
- 5. Lok CE, Huber TS, Lee T, et al. KDOQI clinical practice guideline for vascular access: 2019 update. Am J Kidney Dis 2020; 75 (4 Suppl 2): S1‐S164.
- 6. Nassar GM, Ayus JC. Infectious complications of the hemodialysis access. Kidney Int 2001; 60: 1‐13.
- 7. Ravani P, Palmer SC, Oliver MJ, et al. Associations between hemodialysis access type and clinical outcomes: a systematic review. J Am Soc Nephrol 2013; 24: 465‐473.
- 8. Sahli F, Feidjel R, Laalaoui R. Hemodialysis catheter‐related infection: rates, risk factors and pathogens. J Infect Public Health 2017; 10: 403‐408.
- 9. Lazarus B, Bongetti E, Ling J, et al. Multifaceted quality improvement interventions to prevent hemodialysis catheter‐related bloodstream infections: a systematic review. Am J Kidney Dis 2023; 82: 429‐442.
- 10. Rha B, See I, Dunham L, et al. Vital signs: health disparities in hemodialysis‐associated Staphylococcus aureus bloodstream infections: United States, 2017–2020. MMWR Morb Mortal Wkly Rep 2023; 72: 153‐159.
- 11. Murea M, James KM, Russell GB, et al. Risk of catheter‐related bloodstream infection in elderly patients on hemodialysis. Clin J Am Soc Nephrol 2014; 9: 764‐770.
- 12. Hoen B, Paul‐Dauphin A, Hestin D, Kessler M. EPIBACDIAL: a multicenter prospective study of risk factors for bacteremia in chronic hemodialysis patients. J Am Soc Nephrol 1998; 9: 869‐876.
- 13. Powe NR, Jaar B, Furth SL, et al. Septicemia in dialysis patients: incidence, risk factors, and prognosis. Kidney Int 1999; 55: 1081‐1090.
- 14. Al‐Solaiman Y, Estrada E, Allon M. The spectrum of infections in catheter‐dependent hemodialysis patients. Clin J Am Soc Nephrol 2011; 6: 2247‐2252.
- 15. Phillips J, Chan DT, Chakera A, et al. Haemodialysis vascular catheter‐related blood stream infection: organism types and clinical outcomes. Nephrology (Carlton) 2023; 28: 249‐253.
- 16. Kotwal S, Cass A, Coggan S, et al. REDUCCTION Investigators. Multifaceted intervention to reduce haemodialysis catheter related bloodstream infections: REDUCCTION stepped wedge, cluster randomised trial. BMJ 2022; 377: e069634.
- 17. Kotwal S, Coggan S, McDonald S, et al. Reducing the burden of dialysis catheter complications: a national approach (REDUCCTION): design and baseline results. Kidney360 2020; 1: 746‐754.
- 18. Christensen KL, Holman RC, Steiner CA, et al. Infectious disease hospitalizations in the United States. Clin Infect Dis 2009; 49: 1025‐1035.
- 19. Eisen DP, McBryde ES, Vasanthakumar L, et al. Linking administrative data sets of inpatient infectious diseases diagnoses in far North Queensland: a cohort profile. BMJ Open 2020; 10: e034845.
- 20. Lo Re V, Carbonari DM, Jacob J, et al. Validity of ICD‐10‐CM diagnoses to identify hospitalizations for serious infections among patients treated with biologic therapies. Pharmacoepidemiol Drug Saf 2021; 30: 899‐909.
- 21. Allon M. Dialysis catheter‐related bacteremia: treatment and prophylaxis. Am J Kidney Dis 2004; 44: 779‐791.
- 22. Engemann JJ, Friedman JY, Reed SD, et al. Clinical outcomes and costs due to Staphylococcus aureus bacteremia among patients receiving long‐term hemodialysis. Infect Control Hosp Epidemiol 2005; 26: 534‐539.
- 23. Farrington CA, Allon M. Complications of hemodialysis catheter bloodstream infections: impact of infecting organism. Am J Nephrol 2019; 50: 126‐132.
- 24. Marr KA, Sexton DJ, Conlon PJ, et al. Catheter‐related bacteremia and outcome of attempted catheter salvage in patients undergoing hemodialysis. Ann Intern Med 1997; 127: 275‐280.
- 25. Maya ID, Carlton D, Estrada E, Allon M. Treatment of dialysis catheter‐related Staphylococcus aureus bacteremia with an antibiotic lock: a quality improvement report. Am J Kidney Dis 2007; 50: 289‐295.
- 26. Hayat A, Cho Y, Pascoe EM, et al. Uptake and outcomes of peritoneal dialysis among Aboriginal and Torres Strait Islander people: analysis of registry data. Kidney Int Rep 2024; 9: 1484‐1495.
- 27. Dufour A, Candas V. Ageing and thermal responses during passive heat exposure: sweating and sensory aspects. Eur J Appl Physiol 2007; 100: 19‐26.
- 28. Jean G, Charra B, Chazot C, et al. Risk factor analysis for long‐term tunneled dialysis catheter‐related bacteremias. Nephron 2002; 91: 399‐405.
- 29. Kuehnert MJ, Kruszon‐Moran D, Hill HA, et al. Prevalence of Staphylococcus aureus nasal colonization in the United States, 2001–2002. J Infect Dis 2006; 193: 172‐179.
- 30. Wasik HL, Neu A, Warady B, et al. Standardizing Care to Improve Outcomes in Pediatric End‐stage Kidney Disease (SCOPE) Investigators. The cost of hospitalizations for treatment of hemodialysis catheter‐associated blood stream infections in children: a retrospective cohort study. Pediatr Nephrol 2023; 38: 1915‐1923.
- 31. Pronovost P, Needham D, Berenholtz S, et al. An intervention to decrease catheter‐related bloodstream infections in the ICU. N Engl J Med 2006; 355: 2725‐2732.
- 32. Ştefan G, Stancu S, Căpuşă C, et al. Catheter‐related infections in chronic hemodialysis: a clinical and economic perspective. Int Urol Nephrol 2013; 45: 817‐823.
Abstract
Objectives: To investigate the epidemiology in Australia of catheter‐related infections in a national cohort of adults with kidney failure with incident haemodialysis central venous catheters (CVCs).
Study design: Cohort study nested within a cluster‐randomised trial (REDUcing the burden of dialysis Catheter Complications, REDUCCTION); analysis of prospectively collected trial data, linked with Australian and New Zealand Dialysis and Transplant (ANZDATA) registry and state hospitalisations data.
Setting: Thirty‐four health services in Australia (excluding Western Australia) that provide chronic haemodialysis and participated in the REDUCCTION trial.
Participants: Adults (18 years or older) with chronic kidney failure who received incident haemodialysis CVCs during 20 December 2016 – 31 March 2020.
Main outcome measures: Hospitalisation with any haemodialysis CVC infection; haemodialysis CVC‐related bloodstream infections reported during the trial and verified by an independent panel.
Results: Our analysis included 3943 adults with chronic kidney failure; their mean age was 60.4 years (standard deviation, 15.5 years); 1556 were women (39.5%) and 485 were Aboriginal or Torres Strait Islander people (12.3%). Catheter‐related infections were coded for 644 hospitalisations (24.5 per 100 patient‐years; 95% confidence interval [CI], 22.6–26.4 per 100 patient‐years); the incidence was higher among people under 55 years of age (adjusted incidence rate ratio [IRR], 1.55; 95% CI, 1.21–1.98) and those aged 55–70 years (adjusted IRR, 1.34; 95% CI, 1.05–1.70) than among people over 70 years of age. Community‐onset haemodialysis catheter‐related bloodstream infections were responsible for 159 hospitalisations (8.2% of 1938 infection‐related hospitalisations); 57 of 650 infection‐related hospitalisations of people under 55 years of age (8.8%), 62 of 640 of people aged 55–70 years (9.7%), and 40 of 648 of people over 70 years of age (6.2%). The median length of hospital stay with community‐onset haemodialysis CVC‐related bloodstream infections was ten days (interquartile range, 5–15 days), metastatic spread of infection was detected in twelve cases (7.5%), and four people died in hospital (2.5%); 40 removed haemodialysis CVCs did not require replacement. Nineteen of 121 hospitalisations for which the information was available included intensive care unit admissions (15.7%; median stay, 2.7 days; IQR, 1.1–4.6 days). The risk of haemodialysis CVC‐relatedStaphylococcus aureus bloodstream infection declined with age (relative risk ratio, 0.65 per decade; 95% CI, 0.47–0.89).
Conclusions: The health burden of haemodialysis CVC infections in Australia is substantial, particularly among adults under 70 years of age.