gms | German Medical Science

GMS Health Technology Assessment

Deutsche Agentur für Health Technology Assessment (DAHTA)

ISSN 1861-8863

Clinical effectiveness and cost-effectiveness of central venous catheters treated with Minocycline and Rifampicin in preventing bloodstream infections in intensive care patients

HTA Summary

  • corresponding author Silke Neusser - ISEG – Institut für Sozialmedizin, Epidemiologie und Gesundheitssystemforschung, Witten, Germany
  • author Eva Maria Bitzer - ISEG – Institut für Sozialmedizin, Epidemiologie und Gesundheitssystemforschung, Witten, Germany
  • author Ingeborg Mieth - ISEG – Institut für Sozialmedizin, Epidemiologie und Gesundheitssystemforschung, Witten, Germany
  • author Christian Krauth - Medizinische Hochschule Hannover, Institut für Epidemiologie, Sozialmedizin und Gesundheitssystemforschung, Hannover, Germany

GMS Health Technol Assess 2012;8:Doc08

DOI: 10.3205/hta000106, URN: urn:nbn:de:0183-hta0001062

This is the original version of the article.
The translated version can be found at: http://www.egms.de/de/journals/hta/2012-8/hta000106.shtml

Published: October 18, 2012

© 2012 Neusser et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by-nc-nd/3.0/deed.en). You are free: to Share – to copy, distribute and transmit the work, provided the original author and source are credited.

The complete HTA Report in German language can be found online at: http://portal.dimdi.de/de/hta/hta_berichte/hta329_bericht_de.pdf


Outline

Abstract

The use of central venous catheters coated with antibiotics can avoid bloodstream infections with intensive care patients. This is the result of a scientific examination which has been published by the DIMDI. Costs could be also saved in this way. However, according to the authors, the underlying studies do not allow absolutely valid statements.

Keywords: anti-infective agents, bloodstream infection, catheters, central venous, HTA, intensive care patient, Minocycline, review, Rifampin


Summary

Scientific background

Central venous catheters (CVC) are routinely used in intensive care medicine. They allow long-term and continuous administration of larger amounts of liquids and drugs, in particular infusion liquids which irritate veins. Further, they are used to monitor the central venous pressure. Extrapolations show that there are about 5 million CVC patient days in German intensive care units in 2010.

CVC bear the risk of complications. A substantial risk is CVC-associated bloodstream infection, i.e. bloodstream infection relating to indwelling catheters. In Germany infection rates of 1.3 to 2.1 bloodstream infections per 1,000 CVC days are estimated. This implies 6,400 to 10,600 cases of CVC-associated bloodstream infections in 2010. Migration of skin germs at the injection site (while the CVC is attached or later via the catheter parts located at the outer surface, and subsequent colonisation of the catheter tip) is the most frequent route of infection.

Infection rates are considerably reduced if hygiene rules are observed. In addition, antimicrobially treated CVC are available which promise a further reduction of infection rates. CVC coated with the antibiotics minocycline and rifampin (MR-CVC) are commercially available in Germany. Clinical studies suggest that infection rates are considerably reduced by MR-CVC. However, there are concerns about allergic reactions and a long-term development of resistances.

Research questions

Medical evaluation

Do MR-CVC with a comparable compliance of hygiene requirements contribute:

  • to a lower rate of bloodstream infections than standard CVC in intensive care patients?
  • to a lower rate of bloodstream infections than other antimicrobially treated CVCs in intensive care patients?
Health economic evaluation
  • Are MR-CVC cost-effective compared to alternative CVC (e. g. standard CVC or chlorhexidine, silver-sulfadiazine-coated [CHSS] catheters)?
  • What are the effects on health care costs and mortality of intensive care patients if MR-CVC are used (instead of alternative CVC)?
  • Does the use of MR-CVC result in cost savings (compared to alternative catheters)?

Further, the health economic evaluation includes an examination of the transferability to the German health care context.

Ethical, social and legal aspects
  • Which specific ethical, social and legal implications are to be considered regarding the usage of MR-CVC compared to standard CVC?
  • Do RCT included into the medical evaluation indicate aspects such as allergic reactions to MR-CVC?

Methods

A systematic literature search of 32 databases was conducted in September 2011 to evaluate medical effectiveness, cost-efficiency and to identify ethical, social and legal aspects of the utilization of MR-CVC in preventing bloodstream infections. The search was performed on English, German and French publications.

In a first step the titles and abstracts of identified publications were screened according to predefined selection criteria. In a second step the then selected publications were examined in full text. The selection of the literature was done by two independent reviewers. The methodological quality of included publications was assessed using extraction sheets specifically prepared for each research question.

Medical evaluation

The systematic literature search identified 737 publications. 83 publications were selected to review the full text version for the medical evaluation. Furthermore, search results revealed an additional systematic review. In total, the systematic literature search revealed eight randomised controlled trials (RCT) and 18 systematic reviews. Due to the large number of systematic reviews only the seven latest ones were included for this evaluation. These reviews included the majority of the eight RCT.

Two of eight included RCT, are of good methodological quality (Level of evidence 1+). Another six RCT are flawed by systematic bias (Level of evidence 1-). In two RCT the systematic bias contributes to an underestimation of the protective effects of MR-CVC. In another four RCT systematic bias contributes to an overestimation as well as an underestimation of the protective effects of MR-CVC. All RCT include patients who are at risk or at high risk for infections. However, only three RCT include intensive care patients. One of these RCT is of good quality (level of evidence 1+). The other RCT with a good methodological quality includes intensive care patients only partly.

Despite distinct methodological differences all RCT show a protective effect of MR-CVC compared to standard CVC or other antimicrobially treated CVC. However, this beneficial effect does not reach statistical significance in most of the trials. This is the case for the primary outcome “CVC-related bloodstream infection” and the secondary outcome “CVC colonisation”. Only two RCT address the secondary outcome “mortality caused by CVC-related bloodstream infection”. In both trials this is a rare incident occurring only in the control group.

Only one RCT of good methodological quality compares MR-CVC with standard CVC. This trial includes only intensive care patients, the main target group of this medical evaluation. The relative risk for the outcome “CVC-related bloodstream infection” is 0.53 (0.20 to 1.39). For the outcome “CVC colonisation” the relative risk is 0.43 (0.26 to 0.70). The other studies show even higher protective effects which in some cases are statistically significant. However, these trials have a high risk of bias.

Another RCT of good methodological quality compares MR-CVC with other antimicrobially treated CVC. However, this trial only partly includes intensive care patients. For the control group extraluminally CHHS impregnated CVC were used. For the outcome “CVC-related bloodstream infection” the relative risk is 0.09 (0.01 to 0.68) and for the CVC colonisation the relative risk is 0.38 (0.25 to 0.56).

All seven systematic reviews conducted a metaanalysis including most or all eight RCT. The pooled results suggest a protective effect of MR-CVC.

Discussion

Despite distinct methodological differences and heterogeneous study populations all RCT show a protective effect of MR-CVC compared to standard CVC or other antimicrobially treated CVC. However, this beneficial effect does not reach statistical significance in most of the trials.

Reflecting back on the research questions, the medical evaluation suggests that MR-CVC with a comparable compliance of hygiene requirements contribute:

  • to a lower rate of bloodstream infections than standard CVC in intensive care patients.
  • to a lower rate of bloodstream infections than other antimicrobial treated CVC in intensive care patients.

However, it is remarkable that no RCT has been performed since 2002. The large number of systematic reviews indicates uncertainties which could not be resolved by any of the RCT reviewed for this evaluation. For example, the authors of the systematic reviews specifically discuss unresolved question regarding the emergence of resistance to minocycline or rifampin. Additionally, it has to be noted that all RCT were supported by the manufacturer of MR-CVC, and five RCT were conducted by the same group of authors.

Health economic evaluation

Twelve publications of the 737 articles identified during the first step of the literature search process were selected for a full text review. Five relevant publications were identified: a systematic review, a cohort study, and three decision models. An additional systematic literature search regarding central parameters of the included decision models was conducted to assess the models’ quality. Relevant parameters of the decision models are additional length of stay, additional costs, and attributable mortality due to CVC-associated bloodstream infections. Overall, 13 relevant publications were identified: ten studies cover additional length of stay, eight studies additional costs, and 10 studies focus on attributable mortality of CVC-associated bloodstream infections.

All four studies examining the cost-effectiveness of the intervention “MR-CVC as a means to prevent bloodstream infections” report cost savings due to the use of MR-CVC. Cost savings of 103 to 246 euro are reported.

Two of the decision models are lifetime analyses and consider mortality and quality-adjusted life years (QALY). The models find QALY-gains of 1.6 to 34 QALY per 1,000 patients.

Discussion

The decision models are based on conservative assumptions concerning additional costs of CVC-associated bloodstream infections. However, only the latest model includes the most recent literature on additional costs due to bloodstream infections. The latest publications report considerably lower additional costs (and conversely lower saving potentials by avoiding bloodstream infections) than previous literature.

Study results on attributable mortality of CVC-associated bloodstream infections (which are included in the different decision models) are heterogeneous and report attributable mortality rates between 1.8% and 35.0%. In most studies attributable mortality is not significant. Thus, no clear statements concerning attributable mortality can be derived.

Regarding the health economic questions, it can be stated that

  • MR-coated CVC induce cost savings of about 100 euro per patient.
  • QALY-gains due to MR-coated CVC are small if they exist at all.

The health economic studies prove that the cost savings rise with increasing number of catheter days, which means that the potential for cost savings due to MR-coated CVC is increasing as well.

Ethical, social and legal aspects

The search results were screened for publications on ethical, social and legal aspects in the context of MR-CVC. Additionally, eight RCT included in the medical evaluation were examined for information on allergic reactions and the development of resistance to minocycline or rifampin.

The systematic literature search identified no articles explicitly dealing with ethical, social or legal aspects concerning MR-CVC.

Side effects are addressed by two of the eight RCT. Both RCT report that no allergic reactions have occurred during the trial.

With regard to the kind of organisms usually colonising CVC, the RCT included into the medical evaluation show that: MR-CVC in the intervention group are more prone to a colonisation by Candida species than CVC in the control groups. However, regarding the clinically relevant outcome “CVC-related bloodstream infection” no indication for stronger involvement of Candida spp. was found.

Half of the RCT included into the medical evaluation examine the emergence of resistance to minocycline and rifampin. None of these studies indicated the development of resistance to either agent. However, it should be noted that a rather small number of strains were tested. The risk of the emergence of resistance to minocycline and rifampin is discussed by five of the seven systematic reviews included in the medical evaluation. The authors conclude that more research is needed to clarify this aspect.

Discussion

The problem of side effects and the risk of the emergence of resistance to minocycline and rifampin are only partly addressed in the RCT. Yet, the systematic reviews discuss further requirements of clarification. While in-vitro studies indicate the risk of the development of resistance, the RCT suggest no such risk. However, due to the rather small sample seize the RCT are not able to record rare incidents.

Conclusions

The utilization of MR-CVC shows protective effects and contributes to cost savings. Nevertheless, the application of this technology can not be recommended without limitations. It should be noted that no RCT has been performed since 2002 and that all available RCT were supported by the manufacturer of MR-CVC. Additionally, five of the eight RCT are conducted by the same group of authors. The large number of systematic reviews indicates uncertainties, especially regarding the risk of the development of resistance to minocycline and rifampin. One reason is that in-vitro studies did find indications for the emergence of resistance against these antibiotics.

To clarify the medical effectiveness, a larger RCT conducted without manufacturers support is essential. Questions concerning the risk of allergic reactions and the possible development of antibiotic-resistant strains should be clarified by surveillance.