Tigecycline versus imipenem/cilastatin for complicated intra-abdominal infection: clinical cure at test-of-cure

Registration and administrative information (PROSPERO field set)

Extracted values, and where each came from

One row per extracted value. Every row carries the value, the verbatim sentence it was read from, a resolvable link to the source, and whether the number was read or derived. Where any of those is absent the row says so rather than omitting the value.

Risk of bias (RoB-2)

Variant:

Unit assessed: —

Assessor 1: (anthropic family). Assessor 2: (openai family).

Overall judgement per trial

Inter-assessor agreement

This assessment is DUAL — two assessors from different model families, the second asked blind. The agreement rate is withheld pending adjudication. Both readers were given this review’s decision rule; an earlier version of this sentence said the blinding had withheld it, and that was wrong. What the 15 domain comparison(s) show is a reading difference that runs in OPPOSITE directions by domain — on domains 1 to 3 the second assessor more often declines to judge at all, and on domains 4 and 5 it more often judges low. A single agreement rate averages those two behaviours into one number that describes neither. Both assessors’ per-result judgements are shown in the table above, unadjudicated, and neither is this review’s finding. No adjudication has been performed, so this review holds no final risk-of-bias judgement for these results.

References

Included studies

• — https://clinicaltrials.gov/study/NCT00081744
• — https://clinicaltrials.gov/study/NCT00136201
• — https://clinicaltrials.gov/study/NCT01721408

Regulatory documents

Methods and guidance

Sources

Visual abstract

⬇ SVG (vector) 2 KB ⬇ PNG 4500x2100 323 KB

All raster formats were generated at build time from the same SVG as the graphic above (SHA-256, first 16 hex characters: 6ecf6597e2eb47a9), at 600 dpi for a 6-inch print width, on a white ground. They are not screenshots of the page and do not change with the theme you are reading in.

Submission formats written to the figures directory, not embedded here because an uncompressed 600 dpi TIFF is tens of megabytes: EPS (vector, journal's first choice for line art) — visual-abstract.eps, 4 KB; TIFF (uncompressed, 600 dpi -- journal fallback) — visual-abstract.tiff, 27,685 KB.

Projected from the canonical object, so it carries the same k, the same pooled estimate and the same interval as the paper and cannot drift from them. The interval is drawn to scale against the no-difference line, which it EXCLUDES: 0.888 to 0.984, with no difference at 1.

Risk of bias, both assessors

⬇ SVG (vector) 11 KB

Only the vector format is offered for this figure, by choice: it is line art — discrete marks, glyphs and labels — which SVG reproduces exactly at any size, so a raster would add weight without adding fidelity. The build was able to rasterise; it was asked not to.

Every cell carries BOTH independent cross-family assessments and they are not reconciled: showing one column would be a reconciliation presented as an observation. Glyph as well as colour, so the panel survives greyscale printing and colour-blind reading. A trial with no judgement is shown as a row of not-assessed markers rather than omitted, so a gap in the assessment cannot be mistaken for a clean assessment. Agreement as measured: {'per domain agreed': 10, 'per domain total': 15, 'per domain rate pct': 66.7, 'overall agreed': 0, 'overall total': 3, 'overall is derived': True, 'recomputed from the stored judgements': True, 'why recomputed': 'The stored disagreement tallies were frozen as text and 12 of 22 no longer match the judgements in the object -- a later guidance pass moved 51 the randomisation process judgements and the tallies were never recomputed. This is derived from what the object holds now.'}.

GOSH

⬇ SVG (vector) 1 KB ⬇ PNG 3500x540 80 KB

All raster formats were generated at build time from the same SVG as the graphic above (SHA-256, first 16 hex characters: dd33d4f986f7c7c5), at 600 dpi for a 6-inch print width, on a white ground. They are not screenshots of the page and do not change with the theme you are reading in.

Submission formats written to the figures directory, not embedded here because an uncompressed 600 dpi TIFF is tens of megabytes: EPS (vector, journal's first choice for line art) — gosh.eps, 2 KB; TIFF (uncompressed, 600 dpi -- journal fallback) — gosh.tiff, 5,537 KB.

Deliberately not drawn. Every subset meta-analysis of 3 trials is 7 points; a cloud that small cannot show the multimodality GOSH exists to reveal, and a reader would take the picture as evidence of its absence.

Trial-sequential analysis

⬇ SVG (vector) 1 KB ⬇ PNG 3500x650 87 KB

All raster formats were generated at build time from the same SVG as the graphic above (SHA-256, first 16 hex characters: 25ed9800fdf56614), at 600 dpi for a 6-inch print width, on a white ground. They are not screenshots of the page and do not change with the theme you are reading in.

Submission formats written to the figures directory, not embedded here because an uncompressed 600 dpi TIFF is tens of megabytes: EPS (vector, journal's first choice for line art) — tsa.eps, 2 KB; TIFF (uncompressed, 600 dpi -- journal fallback) — tsa.tiff, 6,665 KB.

TSA boundaries depend entirely on a target information size that must be pre-specified. This review's protocol registers none, so any boundary drawn here would be a parameter chosen after seeing the data -- which is the practice TSA exists to protect against.

Meta-regression

⬇ SVG (vector) 1 KB ⬇ PNG 3500x540 87 KB

All raster formats were generated at build time from the same SVG as the graphic above (SHA-256, first 16 hex characters: 2316b730fa24744e), at 600 dpi for a 6-inch print width, on a white ground. They are not screenshots of the page and do not change with the theme you are reading in.

Submission formats written to the figures directory, not embedded here because an uncompressed 600 dpi TIFF is tens of megabytes: EPS (vector, journal's first choice for line art) — bubble.eps, 2 KB; TIFF (uncompressed, 600 dpi -- journal fallback) — bubble.tiff, 5,537 KB.

The protocol pre-specifies no moderator, and at k = 3 a meta-regression would be fitted on 3 points. Not drawn rather than drawn with a caveat: a bubble plot invites reading a slope, and there is no slope here that any reader should read.

Summary of findings

1. This certainty rating is PENDING, not rated. It rests on a risk-of-bias judgement that this review has not finalised: two assessors read every contributing result independently, they disagree, and no adjudication has been performed, so this review holds no final risk-of-bias judgement for these results. Cochrane rates certainty per outcome by aggregating the risk-of-bias judgements of the results contributing to it, so a certainty rating cannot be more final than the risk-of-bias assessment it reads. The rating this object currently records, and the steps behind it, are shown below as the work so far rather than as this review's answer. No level is published for this outcome -- not the recorded one, not a more cautious one, and not a midpoint -- because choosing one would be the judgement that has not been made.
2. Read this beside `whether the estimand was established`. That flag records that every contributing trial measures the same quantity. It records nothing about what that quantity was measured against -- whether the comparators are of one kind, whether each comparison was randomised, whether it was concurrent. The two questions were never separated. Named as P48 in PAGE-STANDARD.md v1.21.0 and swept by scripts/audit_mixed_contrast_pools.py.

Reproducibility artifact

This page CANNOT be rebuilt from what is recorded here: no canonical object is recorded and no registered commit is recorded. Each figure on the Analysis tab downloads as an SVG carrying exactly the values shown.

Other recorded qualifications

Statements this review records about the limits of what it establishes, which are not shown elsewhere on this page. They are reproduced from the object verbatim and are not summarised.

Paper

Each statement below cites the record it came from; those sources are listed at the end of every section. Where the record holds nothing, the section says so by name rather than being left out, so a procedure that was not carried out can be told from one that simply was never mentioned.

Title

Tigecycline versus imipenem/cilastatin for complicated intra-abdominal infection: clinical cure at test-of-cure¹

Abstract

Question. Among trials randomising adults with complicated intra-abdominal infection to tigecycline or to imipenem/cilastatin, how does the proportion cured at the test-of-cure visit compare, and does that comparison survive the choices made in reaching it?¹

Methods. estimates were pooled under a random-effects model with the REML estimator; risk of bias was assessed with RoB 2 (Cochrane risk-of-bias tool for randomized trials); and certainty was rated with GRADE, following the Cochrane Handbook chapter 14. Randomised evidence starts HIGH and is rated down with reasons.²

Results. Clinical cure at the test-of-cure visit: risk ratio 0.935 (0.888 to 0.984) across 3 trials; 1 of these pooled outcome(s) carries a stated qualification -- a referral or a recorded finding -- given in full in Results and not reproduced here; and no statistical heterogeneity was detected (I-squared 7.3%).³

Introduction

Background. This review asks: Among trials randomising adults with complicated intra-abdominal infection to tigecycline or to imipenem/cilastatin, how does the proportion cured at the test-of-cure visit compare, and does that comparison survive the choices made in reaching it? It pools 3 randomised trials, against imipenem/cilastatin. The outcome pooled is Clinical cure at the test-of-cure visit.¹

Methods — search

Primary trials. No bibliographic search for primary trials was run. The 3 included trials were identified by reading named registrations on the dates recorded on each entry. Published syntheses. PubMed E-utilities esearch + esummary, executed 2026-08-21. Query as executed: (tigecycline[tiab]) and ("intra-abdominal"[tiab] OR cIAI[tiab]) and (meta-analysis[pt] OR "systematic review"[pt] OR meta-analysis[tiab] OR "pooled"[tiab]). It matched 22 record(s), of which 22 were retrieved and 22 read; 1 appraised.¹

Methods — synthesis

For clinical cure at the test-of-cure visit, a random model was fitted with the REML estimator.¹

Intervals are 95% confidence intervals.²

Methodological decisions follow Higgins JPT, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, et al (editors). Cochrane Handbook for Systematic Reviews of Interventions version 6.5.1. Cochrane, 2025., version 6.5.1, and the sections relied on are listed in the object rather than cited generically.³

Risk of bias was assessed with RoB 2 (Cochrane risk-of-bias tool for randomized trials) , following . 3 result-level assessments were made.⁴

A ceiling applies to every risk-of-bias judgement in this review; it is stated with the assessment itself rather than summarised here.⁵

Certainty of evidence was rated with GRADE, following the Cochrane Handbook chapter 14. Randomised evidence starts HIGH and is rated down with reasons., following . 1 pooled outcome(s) were rated. No domain is rated UP; these are randomised trials starting HIGH.⁶

Results

The pooled risk ratio for clinical cure at the test-of-cure visit was 0.935 (0.888 to 0.984, 95% interval) across 3 trials, against imipenem/cilastatin.¹

Noted on this pool. A published network meta-analysis of fifteen studies and 6,745 participants concludes: 'No differences in clinical and microbiological outcomes were observed between different carbapenems and TGC' (Yu et al., Medicine 2019, PMID 31577763). This pool reports rr 0.9351 (0.8885 to 0.9842) on three trials -- an interval that excludes no difference, read as tigecycline being inferior on clinical cure. Read this beside the estimate. This review pools cure rates for tigecycline in complicated intra-abdominal infection and says nothing, anywhere, about all-cause mortality. Tigecycline carries an FDA boxed warning for increased all-cause mortality, and complicated intra-abdominal infection is among the indications it was drawn . Independently of the label, one of the three pooled registrations posts an asymmetry of its own: NCT01721408 reports serious adverse events in 32 of 232 on tigecycline against 16 of 232 on imipenem/cilastatin, read from its own posted results. A cure-rate estimate for this drug, in this indication, presented without that, is the smaller half of the story. The analysis population is part of the estimand and this pool crosses it. Read from the registrations on 2026-08-21: NCT00081744 registers 'no primary outcome is recorded on this registration' (population: None recorded); NCT00136201 registers 'Clinical response for all microbiologically evaluable and microbiologically modified intent-to-treat subjects' (population: ME and m-mITT); NCT01721408 registers 'Clinical Response at the Test-of-Cure (TOC) Assessment Within the Clinically Evaluable (CE) Population' (population: CE). The populations pooled are CE and ME and m-mITT. A cross-lane check reported that this pool carries three of four trials and that the omitted one is the only one favouring the drug. That omitted trial is Study 3074A1-306 (tigecycline vs imipenem/cilastatin, the SAME comparator as this pool), excluded here on a reason now RETRACTED: this object had claimed its per-arm counts exist only inside an unopenable pooled report, but a standalone publication (clinical cure 242/265) and the FDA label report both arms. NCT00230971 (ceftriaxone plus metronidazole) is a DIFFERENT, separately-excluded trial and must not be confused with 306. Recomputing the pool with 306 gives k=4 RR 0.9591 (0.9040 to 1.0175), which includes no difference, and leave-one-out on that k=4 set omitting 306 returns exactly the 0.9351 shown above — so the inferiority reading is an artefact of omitting a known-eligible trial and is WITHDRAWN. 306 is NOT added to the pool here: its counts are pending source-of-record verification, and its identifier is unresolved (it carries no ClinicalTrials.gov registration). The omission is real regardless of which identifier it carries. The inferiority conclusion does not survive this object's own adjustment. The Hartung-Knapp interval at k = 3 is 0.8327 to 1.0501 on t = 4.3027 with 2 degrees of freedom, and it includes no difference. A reader taking inferiority from this page is reading the unadjusted interval only. A clinically evaluable population excludes protocol violators and indeterminate responses, and microbiologically-defined populations exclude patients without a qualifying pathogen. Those are selected subsets of the modified intent-to-treat population and systematically yield higher cure rates. A cure-rate ratio computed in one is not an estimate of the same quantity as one computed in another, so this pooled value depends on which trial registered which population. One contributing trial is OPEN-LABEL (NCT00136201 registers masking none) on an assessed clinical-response endpoint; one registers no primary outcome at all (NCT00081744, open item O0a) so its result cannot be checked against anything; and the three define their primary in different analysis populations. GRADE certainty is very LOW. This object's own sensitivity analyses are the place to read this estimate . Changing the summary statistic to an odds ratio gives 0.4987 (0.1955 to 1.2722) and reversing which category counts as the event gives 1.8683 (0.7784 to 4.4844) -- both include no difference. An inferiority reading taken from 0.9351 (0.8885 to 0.9842) does not survive this object's own robustness checks, and that is recorded here rather than left for a reader to assemble. Not that the arithmetic is wrong. Not that the direction would reverse -- that would need the per-population counts and is not established. And not that every contributing registration establishes a non-inferiority design: NCT01721408's posted results include nonInferiorityType NON_INFERIORITY_OR_EQUIVALENCE and a -15% non-inferiority criterion, while the other contributing registrations do not settle that design question here. Which twelve studies the network carries that this object does not. The abstract names none and no included-study table was read, so the gap is counted and not identified. Read this before the interval above. The estimate is 0.9351 (0.8885 to 0.9842), which excludes no difference and reads as tigecycline being inferior on clinical cure. This project's own hartung-knapp interval at k = 3 IS 0.8327 to 1.0501, on a t critical value of 4.3027 with 2 degrees of freedom, and it includes no difference. The house rule already requires the adjusted interval to be shown beside the unadjusted one where k is small, because that is the honest width; on this pool the two disagree about the conclusion. Yu et al., Medicine 2019, PMID 31577763, a Bayesian network meta-analysis of fifteen studies and 6,745 participants, concluded: 'No differences in clinical and microbiological outcomes were observed between different carbapenems and TGC.' the adjusted interval and the larger literature agree with each other and not with the unadjusted reading. The stored estimate is not changed. Withdrawing or restating a published number is a content decision. What has been done is to put the adjusted interval and the disagreement where a reader meets the number, because a page that concludes more than its own stated method supports had been passing while pages that refuse honestly are marked incomplete.²

The trials' results were closely consistent with one another: I-squared, the share of the variation between them that is more than chance alone would produce, was 7.29%; the estimated variance of the true effects between trials (tau-squared) was 2.5e-05, the heterogeneity test statistic Q was 2.16, on degrees of freedom 2. tau-squared and I-squared are recorded beside this line, on the measure actually used. They are not the grounds on which that measure was chosen.³

The grounds for pooling are recorded as: Identical across the pooled trials: indication, comparator and outcome definition. Analysis population is not identical: the finding block for this pool records the registered populations as none recorded, ME and m-mITT, and CE. Not identical, and pooled anyway with the reason recorded against each: analysis population, blinding, region and trial size. Each difference that could plausibly drive the result has a matching entry in the sensitivity block. Where a later pool on this page rests on the same grounds, they are not restated.⁴

The full model output for every pooled outcome -- the call, the estimator, the heterogeneity statistics and the back-transformed interval -- is stored verbatim on this review object at `results.by_outcome.<outcome>.r_output.verbatim`. It is not reproduced as an Extended data section in this article; an earlier version of this sentence said it was, and no such section exists.⁵

Risk of bias in the included results

Risk of bias was assessed with RoB 2 at the level of each reported result, by two assessors reading independently. Their judgements differ on 3 of the 3 results and no adjudication has been performed, so this review holds no final risk-of-bias judgement for these results. Assessor 1 recorded: 3 at high risk of bias. Assessor 2 recorded: 2 at some concerns, 1 at no information. Both are shown in the risk-of-bias table; neither is the review's finding. The per-domain judgements behind them are recorded with the extracted data.¹

Certainty of the evidence

Certainty was rated with GRADE, following the Cochrane Handbook chapter 14. Randomised evidence starts HIGH and is rated down with reasons..¹

No domain is rated UP; these are randomised trials starting HIGH.²

This certainty rating is pending, not rated. It rests on a risk-of-bias judgement that this review has not finalised: two assessors read every contributing result independently, they disagree, and no adjudication has been performed, so this review holds no final risk-of-bias judgement for these results. Cochrane rates certainty per outcome by aggregating the risk-of-bias judgements of the results contributing to it, so a certainty rating cannot be more final than the risk-of-bias assessment it reads. The rating this object currently records, and the steps behind it, are shown below as the work so far rather than as this review's answer. No level is published for this outcome -- not the recorded one, not a more cautious one, and not a midpoint -- because choosing one would be the judgement that has not been made.³

No final certainty level is published for this outcome. The stored work so far records an open-label trial contributing to a pooled assessed endpoint, one contributing trial with no registered primary outcome, a pool crossing analysis populations, and an inferiority conclusion that does not survive the Hartung-Knapp adjustment at k = 3.⁴

Where a result is rated worse than LOW overall, the domain carrying that judgement is: measurement of the outcome on 3 result(s). The overall rating is the worst domain, so these are the domains that determine it.⁵

Comparison with published syntheses

Published syntheses were identified as follows. PubMed E-utilities, executed 2026-08-21. Query, counts and per-record disposition in ssot/tigecycline-ciai/appraisal/the screen of published syntheses.json.¹

Limitations

Eligible trials that do not contribute: This registrational trial meets every eligibility criterion and is not excluded. Its per-arm figures appear only inside a report that pools it with a trial this object already holds separately, and that report is behind a paywall. No paywall was bypassed.¹

4 claim(s) previously made about this review were corrected and the corrections are retained on the object rather than overwritten.²

What this review does not give a clinician: any absolute effect or number needed to treat and any measure of harm. A relative effect cannot be turned into the benefit or the risk facing one patient without them, so this review can support a judgement about what these trials found and not a judgement about what to do.³

Data availability

The underlying records are the 9 source(s) listed under References; each names the layer it was read at, so a reader can tell a registry record from a published report.¹

Software availability

Intervals are reported at the 95% level.¹

Note on registration

No PROSPERO registration or protocol record is held on this object. That is not the same as knowing this review was not registered, and neither claim is made. It is recorded here as an absence so that a reader does not have to infer it from silence.¹

References

Figures

Trial characteristics

Not reported in this record

The items below were not written because the record does not hold what they would be composed from. They are named here so that an absence is not mistaken for an omission.