How we appraise
Sources are read in full where the full text can be obtained legally (1000 of 1009). The other 9 are listed from the citation or abstract only, labelled "Citation only", and not appraised. Each source read in full is critically appraised with the instrument its design calls for, and filed under each topic it bears on. Sources that do not support a treatment are kept and shown: an evidence library is most useful where it says what the evidence does not show. The method is the same as the NPWT Evidence Library's.
Where the sources come from
The maintainer's reference library (lymphoedema and lipoedema collections, and lymphoedema items filed elsewhere); a publication list supplied by a South African lymphoedema therapist and researcher; legal open-access copies (the Lymphology open archive, Unpaywall, PubMed Central open-access text, SciELO South Africa); and papers retrieved by the maintainer. Paywalls and publisher bot checks are never bypassed. A duplicate check on DOI, PMID and title runs before any source is added.
| Study type | Sources |
|---|---|
| Primary study | 369 |
| Case report | 163 |
| Narrative review | 145 |
| Randomised trial | 104 |
| Systematic review | 99 |
| Diagnostic or measurement study | 42 |
| Guideline or consensus | 32 |
| Textbook | 18 |
| Cohort study | 16 |
| Manufacturer material | 5 |
| Patient information | 5 |
| Magazine article | 3 |
| Conference item | 3 |
| Book chapter | 1 |
| Advertisement | 1 |
| South African guidance | 1 |
| Flagged document | 1 |
| News item | 1 |
Ratings
| Rating | Question it answers | Scale |
|---|---|---|
| GRADE ⊕⊕⊕◯ Moderate | Evidence certainty. How sure can we be of this source's main finding? | GRADE: High ⊕⊕⊕⊕, Moderate ⊕⊕⊕◯, Low ⊕⊕◯◯, Very low ⊕◯◯◯. The card shows the certainty of the source's first-listed (main) finding; every finding and the reason for each downgrade are under "Appraisal and reference". Where GRADE does not apply (opinion, case reports, patient information), the OCEBM level of evidence is shown instead. |
| Guideline quality 3/7 | Guideline quality. How well was this guideline or consensus developed? | AGREE II overall score out of 7, from this library's own appraisal, with the six domain scores in the appraisal. |
Other labels on a card
| Study type | The design: randomised trial, systematic review, cohort, guideline, case report, and so on. Manufacturer material, advertisements and patient leaflets are labelled as such and ranked lowest. |
| Direction | Set by one mechanical rule: the pre-specified primary outcome decides. Positive primary outcome: the treatment or test studied was significantly better than its comparator. Null or negative primary outcome: no significant difference, or harm. Context: no comparison (guidelines, reviews without pooled effects, case reports, measurement studies, opinion), a superseded interim report, or a risk-factor analysis. |
| Figures | Figures checked: every number on the card was compared with the paper's tables and recalculated where a claim depends on it. Citation only: the full text could not be obtained; the source is listed but not appraised. These checks, like the appraisals, are carried out by Anthropic's Claude; they are not independent double extraction. |
| Reviewed | The date the appraisal was last reviewed. |
| No DOI or PMID | 254 sources (mostly guidance documents, wound-care magazines and older South African journals) have no DOI or PMID. Their full citation is given so they can be traced. |
Instruments
RoB 2 for randomised trials; AMSTAR 2 for systematic reviews; AGREE II for guidelines and consensus documents; QUADAS-2 for diagnostic accuracy; Newcastle–Ottawa or ROBINS-I for cohorts; JBI checklists for case reports, case series and text-and-opinion documents. GRADE certainty is given per outcome with a written reason for each downgrade. Funding, conflicts of interest, retraction status and internal errors are recorded for every source.
Changes
- 31 October 2026. Corrections after an external audit: the EFforT-BCRL figure now gives the MLD v placebo comparison (−0.2 percentage points, −2.1 to 1.8) and states that the trial was in chronic BCRL; exercise advice separated from the general 150-minute guideline; contradictory statements on heat, garment evidence and arm precautions reconciled, with certainty stated; a "For the therapist" box added to every topic; sources without a DOI or PMID labelled; a printable list of all sources added; this page now gives how many sources were read in full.
- 19 October 2026. All 18 textbooks read at chapter level; each book page now carries reading notes, and textbook points that bear on practice were added to the relevant topics.
- 19 October 2026. Each topic page rewritten as one collated synthesis by theme; the dated update notes were merged into it and no source was dropped.
- 9 October 2026. Library published. All 1009 sources were re-appraised to the NPWT library standard on 8–9 October 2026; positions that changed on re-appraisal are marked "revised" with the date on the topic pages.
About this library
Maintained by Dr Gregory Weir, vascular surgeon, Pretoria. Who decides what: Dr Weir adds sources to the library. The appraisals, figure checks, direction labels and certainty ratings are made by Anthropic's Claude under the rules on this page. Limitation: there is a single appraiser and no independent double extraction; ratings have not yet been reviewed by a lymphoedema therapist. Therapists who would like to review topics, or who find an error, are invited to write to the address below; every correction is checked against the original paper and acknowledged here.
Funding and conflicts of interest: Dr Gregory Weir is a vascular surgeon in independent private practice who encourages optimal conservative management of lymphoedema. His practice refers patients to lymphoedema therapists and has no direct financial interest in the treatment or surgery of these patients. The practice facilitates the clinical assessment and diagnosis of patients with appropriate referral. The practice has no support from industry.
Privacy
The library collects no personal information through its pages. The site counts visits with Google Analytics, which processes technical data such as IP address and browser type and sets cookies. Privacy notice. Nothing a visitor types into a search or filter is sent anywhere: filtering runs in the browser.
Product names are trademarks of their owners and are used only to identify the products studied. This library has no connection with any manufacturer.
Corrections and suggestions are welcome: please email lymph@vascular.co.za.