CRD summary

This review concluded that the association of tigecycline with excess deaths and non-cure included indications for which it was approved and marketed; tigecycline cannot be relied on in serious infections. The conclusions seem overly strong given the limitations of the review and the available evidence.

Authors' objectives

To assess the safety and efficacy of tigecycline compared to other antibiotics.

Searching

PubMed, EMBASE, Scopus, Web of Science, ClinicalTrials.gov, Cochrane Central Register of Controlled Trials (CENTRAL), Micromedex, Gateways to Clinical Trials, DARE and Current Controlled Trials were searched to April 2011 for studies in English; search terms were reported.

Study selection

Randomised controlled trials (RCTs) that compared treatment with tigecycline to treatment with an antibiotic and reported cure rates and mortality were eligible for inclusion. The included studies recruited patients with a range of infections: intra-abdominal, skin, community-acquired or hospital-acquired pneumonia, diabetic foot infection, methicillin-resistant Staphylococcus aureus and vancomycin-resistant Enterococcus. Most studies administered a single dose of tigecycline 100mg followed by 50mg every 12 hours. Comparator regimens varied substantially.

Two reviewers selected studies for the review; disagreements were resolved by referral to a third reviewer.

Assessment of study quality

There did not appear to be a systematic assessment of study quality; blinding was mentioned.

Data extraction

Data were extracted on cure rates and mortality. Risk differences (RD) with 95% confidence intervals (CI) were calculated.

The authors did not state how many reviewers extracted data.

Methods of synthesis

Pooled risk differences with 95% CI were calculated using a random-effects model; a modified intention-to-treat (ITT) analysis was used which excluded patients who had protocol violations or did not receive at least one dose. Meta-regression was used to investigate confounders; the variables investigated were not prespecified in the methods. Publication bias was evaluated using funnel plots and the Egger’s test.

Results of the review

Thirteen studies (7,434 patients, range 172 to 1,061) met inclusion criteria. Nine studies were described as double blind and four were open-label. Study follow-up ranged from seven to 92 days.

Across the 13 studies, 148 out of 3,788 (4%) patients on tigecycline and 106 out of 3,646 (3%) patients treated with comparator antibiotics died (RD 0.7%, 95% CI 0.1 to 1.2; p=0.01; Ι²=0%). The non-cure rate with tigecycline was also increased (RD 2.9%, 95% CI 0.6 to 5.2; p=0.01; Ι²=25%). There was no evidence of publication bias.

Results of subgroup analyses by type of infection, type of data available for analysis and approval status were reported. Sample size and use of blinding were investigated using meta-regression.

Authors' conclusions

The association of tigecycline with excess deaths and non-cure includes indications for which it is approved and marketed; tigecycline cannot be relied on in serious infections.

CRD commentary

The review addressed a clear review question with reproducible inclusion criteria. Relevant sources were searched. Unpublished studies were included. Only studies in English were included so language bias could not be ruled out. Study selection was conducted in duplicate; it was unclear whether similar methods were used to reduce error during data extraction. It did not appear that study quality was systematically assessed so it was unclear whether the included studies were subject to bias. Some of the trials were described as open label; this was unlikely to result in bias for outcomes such as mortality and outcomes where there may be some level of subjectivity could be affected.

There was homogeneity across the included studies in terms of the intervention evaluated but there was substantial variability in the comparators. Pooling of these trials made the assumption that the efficacy of these comparators – and therefore the relative efficacy of tigecycline – were comparable and this may not be the case. The impact of differences in comparators was not investigated. Therefore, reliability and generalisability of the pooled estimates were uncertain. Populations varied considerably across the studies; the type of infection was investigated in subgroup analyses. The reduction in non-cure rates was statistically significant but the absolute reduction was small (2.9%) and it is unclear whether this is clinically significant.

The conclusions seem overly strong given the limitations of the review and the available evidence.

Implications of the review for practice and research

Practice: The authors stated that tigecycline should not be used when other effective antibiotic choices are available. Use of tigecycline in life-threatening infections for which there are few or no alternative agents may be justifiable but is supported only by anecdotal evidence.

Research: The authors did not state any implications for research.

Funding

National Institutes of Health, USA.

Bibliographic details

Prasad P, Sun J, Danner RL, Natanson C. Excess deaths associated with tigecycline after approval based on noninferiority trials. Clinical Infectious Diseases 2012; 54(12): 1699-1709. [PMC free article: PMC3404716] [PubMed: 22467668]

Original Paper URL

http://cid.oxfordjournals.org/content/54/12/1699.abstract

Indexing Status

Subject indexing assigned by NLM

MeSH

Anti-Bacterial Agents /administration & dosage /adverse effects; Drug Approval; Humans; Minocycline /administration & dosage /adverse effects /analogs & derivatives; Randomized Controlled Trials as Topic; Survival Analysis; Treatment Failure; United States; United States Food and Drug Administration

AccessionNumber

12012027829

Database entry date

12/07/2013

Record Status

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.