Outcomes associated with early removal versus retention of peripherally inserted central catheters after diagnosis of catheter-associated infections in neonates

Contemporary Trends in Global Mortality of Sepsis Among Young Infants Less Than 90 Days: A Systematic Review and Meta-Analysis

Objective

Current knowledge on the global burden of infant sepsis is limited to population-level data. We aimed to summarize global case fatality rates (CFRs) of young infants with sepsis, stratified by gross national income (GNI) status and patient-level risk factors.

Methods

We performed a systematic review and meta-analysis on CFRs among young infants < 90 days with sepsis. We searched PubMed, Cochrane Central, Embase, and Web of Science for studies published between January 2010 and September 2019. We obtained pooled CFRs estimates using the random effects model. We performed a univariate analysis at patient-level and a meta-regression to study the associations of gestational age, birth weight, onset of sepsis, GNI, age group and culture-proven sepsis with CFRs.

Results

The search yielded 6314 publications, of which 240 studies (N = 437,796 patients) from 77 countries were included. Of 240 studies, 99 were conducted in high-income countries, 44 in upper-middle-income countries, 82 in lower-middle-income countries, 6 in low-income countries and 9 in multiple income-level countries. Overall pooled CFR was 18% (95% CI, 17-19%). The CFR was highest for low-income countries [25% (95% CI, 7-43%)], followed by lower-middle [25% (95% CI, 7-43%)], upper-middle [21% (95% CI, 18-24%)] and lowest for high-income countries [12% (95% CI, 11-13%)]. Factors associated with high CFRs included prematurity, low birth weight, age less than 28 days, early onset sepsis, hospital acquired infections and sepsis in middle- and low-income countries. Study setting in middle-income countries was an independent predictor of high CFRs. We found a widening disparity in CFRs between countries of different GNI over time.

Conclusion

Young infant sepsis remains a major global health challenge. The widening disparity in young infant sepsis CFRs between GNI groups underscore the need to channel greater resources especially to the lower income regions.

Systematic Review Registration

[www.crd.york.ac.uk/prospero], identifier [CRD42020164321].

Unnecessary removal of vascular access devices due to suspected infection in Australian intensive care units

BACKGROUND

Vascular access devices suspected of infection are often removed unnecessarily and frequently require replacement. The aim of this study was to identify the prevalence and economic impact of premature, unnecessary device removal due to suspected infection in adult patients admitted to the intensive care unit.

METHODS

Secondary data analysis of a prospectively collected data set detailing central venous catheters and peripheral arterial catheters in 1458 adult patients was conducted in nine Australian intensive care units. Data extracted from the parent database included patient demographics, device, and infection-specific data including the reason for device removal. Cost estimates were based on a recently published review of device utilisation and associated costs in Queensland public hospitals.

RESULTS

A total of 6144 central venous catheter days and 4696 arterial catheter days were studied. Median device dwell time was 7.2 (interquartile range: 5.6-9.0) days for central venous catheters and 6.5 (interquartile range: 4.8-8.5) days for arterial catheters. Device removal due to suspected infection occurred at a rate of 25.7 and 15.3 episodes/1000 catheter days in central venous and arterial catheters, respectively. Central venous and arterial catheter-related bloodstream infections occurred at a rate of 1.8 and 0.2 episodes/1000 catheter days, respectively. Central line-associated bloodstream infection occurred at a rate of 3.3 episodes/1000 catheter days. Local central venous and arterial catheter infection occurred at 0.16 and 0.02 episodes/1000 catheter days, respectively. The difference in incidence between devices suspected of infection and those responsible for infection resulted in AUD$67,087 unnecessarily spent on device replacement.

CONCLUSIONS

Unnecessary device removal due to suspected infection presents a substantial clinical problem which is costly for the healthcare organisation and time-consuming for clinicians and places the patient at an increased risk of iatrogenic complications. There is a need for robust evidence and clinical practice guidelines to inform clinical decision-making to reduce unnecessary device removal.

Microbial epidemiology and risk factors for relapse in gram-negative bacteria catheter-related bloodstream infection with a pilot prospective study in patients with catheter removal receiving short-duration of antibiotic therapy

Background

Infectious Diseases Society of America (IDSA) guidelines suggest 7–14 days’ duration of antibiotic treatment for uncomplicated Gram-negative bacteria (GNB) catheter-related bloodstream infection (CRBSI). The objectives of this study were to review microbial epidemiology, to determine rate and risk factors for relapse, and to compare clinical outcomes in patients receiving long- versus short-duration antibiotic therapy.

Methods

A retrospective phase 1 study was conducted between January 2010 and October 2016 to review microbial epidemiology and to determine the incidence of and risk factors for relapse in patients with GNB CRBSI, according to the IDSA guidelines diagnostic criteria. In phase 2 of the study, patients without risk factors for relapse between November 2016 and October 2017 were prospectively recruited to receive antibiotic therapy for 7 days after catheter removal. Matched patients from the retrospective phase 1 study who had received antibiotic therapy for ≥14 days were selected as a phase 2 control group to compare outcomes.

Results

In phase 1, three most common pathogens identified among 174 cases were Pseudomonas aeruginosa (22.0%), Klebsiella pneumoniae (16.7%), and Stenotrophomonas maltophilia (13.4%). Eighty-nine episodes of infection occurred while patients were receiving antibiotic therapy. Of 140 cases, the relapse rate was 6.4%. Catheter retention was the only risk factor strongly associated with relapse (odds ratio = 145.32; 95% confidence interval 12.66–1667.37, P < 0.001). In phase 2, 11 patients with catheter removal were prospectively recruited to receive short-duration therapy. The number of patients with relapse receiving long- or short-duration therapy was 1 (3%) and 0 (0%), respectively (P = 1.000).

Conclusions

For the management of patients with uncomplicated GNB CRBSI, empiric broad-spectrum antibiotic therapy with adequate coverage of P. aeruginosa should be chosen. Catheter removal should be performed to prevent relapse and shortening the duration of treatment could be considered.

Trial registration

Thai Clinical Trial Registry: TCTR20190914001. Retrospectively registered on 13 September 2019.

Clinical impact of delayed catheter removal for patients with central-venous-catheter-related Gram-negative bacteraemia

BACKGROUND

Gram-negative bacteria are increasingly the cause of catheter-related bloodstream infection (CRBSI), and the prevalence of multi-drug-resistant strains is rising rapidly. This study evaluated the impact of delayed central venous catheter (CVC) removal on clinical outcomes in patients with Gram-negative CRBSI.

METHODS

Between January 2007 and December 2016, patients with Gram-negative bacteraemia and CVC placement, from two tertiary care hospitals, were included retrospectively. Cases with CVC removal more than three days after onset of bacteraemia or without CVC removal were classified as having delayed CVC removal.

RESULTS

In total, 112 patients were included. Of these, 78 had CRBSI (43 definite and 35 probable) and 34 had Gram-negative bacteraemia from another source (non-CRBSI). Enterobacteriaceae were less common pathogens in patients with CRBSI than in patients with non-CRBSI (11.5% vs 41.3%; P<0.001). Delayed CVC removal was associated with increased 30-day mortality (40.5% vs 11.8%; P=0.01) in patients with Gram-negative CRBSI; this was not seen in patients with non-CRBSI (25.0% vs 14.3%; P>0.99). Delayed CVC removal [odds ratio (OR) 6.8], multi-drug-resistant (MDR) Gram-negative bacteraemia (OR 6.3) and chronic renal failure (OR 11.1) were associated with 30-day mortality in patients with CRBSI. The protective effect of early CVC removal on mortality was evident in the MDR group (48.3% vs 18.2%; P=0.03), but not in the non-MDR group (11.1% vs 0%; P=0.43).

CONCLUSION

CVCs should be removed early to improve clinical outcomes in patients with Gram-negative CRBSI, especially in settings where MDR isolates are prevalent.