The diagnostic performance of novel skin-based in-vivo tests for tuberculosis infection compared with purified protein derivative tuberculin skin tests and blood-based in vitro interferon-γ release assays: a systematic review and meta-analysis
Bibliographic record
Abstract
BACKGROUND: Novel skin-based tests for tuberculosis infection might present suitable alternatives to current tests; however, diagnostic performance of new tests compared with the purified protein derivative-tuberculin skin test (TST) or interferon-γ release assays (IGRA) needs systematic assessment. METHODS: In this systematic review and meta-analysis, we searched English (Medline OVID), Chinese (Chinese Biomedical Literature Database and the China National Knowledge Infrastructure), and Russian (e-library) databases from the inception of each database to May 15, 2019, (with updated search of the Russian and English databases on Oct, 20 2020) using terms "ESAT6" OR "CFP10" AND "skin test" AND "Tuberculosis" OR "C-Tb" OR "Diaskintest". We included studies reporting on the performance of index tests alone or compared with a comparator. Inclusion criteria varied according to review objectives and performance outcome, but reporting of test cut-offs for positivity applied to study population was required from all studies. We used a hierarchy of reference standards for tuberculosis infection consistent with the 2020 WHO framework to evaluate diagnostic performance. Two authors independently reviewed the titles and abstracts for English and Chinese (LF and MK) and Russian studies (MK and VN). Study quality was assessed with QUADAS-2. Pooled random-effects estimates are presented when appropriate for total agreement proportion, sensitivity in microbiologically confirmed tuberculosis and specificity in cohorts with low risk of tuberculosis infection. This study is registered with PROSPERO, CRD42019135572. FINDINGS: We identified 1466 original articles, of which 37 (2·5%) studies, including 10 915 individuals (7111 Diaskintest, 2744 C-Tb, 887 EC, 173 DPPD), were included in the qualitative analysis (29 [78%] studies of Diaskintest, five [15%] studies of C-Tb, two [5%] studies of EC-skintest, and one [3%] study of DPPD). 22 (1·5%) studies including 5810 individuals (3143 Diaskintest, 2129 C-Tb, 538 EC-skintest) were included in the quantitative analysis: 15 (68%) of Diaskintest, five (23%) of C-Tb, and two (9%) of EC-skintest. Tested sub-populations included individuals with HIV, children (0-18 years), and individuals exposed to tuberculosis. Studies were heterogeneous with moderate to high risk of bias. Nine head-to-head studies of index test versus TST and IGRA permitted direct comparisons and pooling. In a mixed cohort of people with and without tuberculosis, Diaskintest pooled agreement with IGRA was 87·16% (95% CI 79·47-92·24) and 55·45% (46·08-64·45) with TST-5 mm cut-off (TST5 mm). Diaskintest sensitivity was 91·18% (95% CI 81·72-95·98) compared with 88·24% (78·20-94·01) for TST5 mm, 89·66 (78·83-95·28) for IGRA QuantiFERON, and 90·91% (79·95-96·16) for TSPOT.TB. C-Tb agreement with IGRA in individuals with active tuberculosis was 79·80% (95% CI 76·10-83·07) compared with 78·92% (74·65-82·63) for TST5 mm/15 mm cut-off (TST5 mm/15 mm). TST5/15mm reflects threshold in cohorts that applied stratified cutoffs: 5 mm for HIV-infected, immunocompromised, or BCG-naive individuals, and 15mm for BCG-vaccinated immunocompetent individuals. C-Tb sensitivity was 74·52% (95% CI 70·39-78·25) compared with a sensitivity of 78·18% (67·75-85·94) for TST5 mm/15 mm, and 71·67% (63·44-78·68) for IGRA. Specificity was 97·85% (95% CI 93·96-99·25) for C-Tb versus 93·31% (90·22-95·48) for TST 15 mm cut-off and 99·15% (79·66-99·97) for IGRA. EC-skintest sensitivity was 86·06% (95% CI 82·39-89·07). INTERPRETATION: Novel skin-based tests for tuberculosis infection appear to perform similarly to IGRA or TST; however, study quality varied. Evaluation of test performance, patient-important outcomes, and diagnostic use in current clinical algorithms will inform implementation in key populations. FUNDING: StopTB (New Diagnostics Working Group) and FIND. TRANSLATIONS: For the Chinese and Russian translations of the abstract see Supplementary Materials section.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.007 | 0.001 |
| Bibliometrics | 0.001 | 0.003 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".