Read enough about lymphoedema or lipoedema and you will meet results that seem to pull in different directions. One review reports that a complication is common, another reports a much lower figure. One trial finds a volume difference, the next finds little. A guideline says evidence is limited while a clinic page sounds confident. Before concluding that "the research is a mess", it helps to ask a more precise question: what exactly differs between these sources? Often the answer is the question each one asked, not the biology it studied.
Editorial note: this article is educational and has not been medically reviewed. It teaches a way of reading sources. It does not diagnose, rank treatments, or tell any reader what to do about their own health.
Disagreement is not always contradiction
Two studies can both be careful and still report different numbers because they measured different things in different people at different times. The skill is to separate a real conflict, where comparable studies on the same question disagree, from a difference in design that fully explains the gap. The framework below uses six questions. It builds on the PICO-S structure used across INNERSTANDIN (Population, Intervention or exposure, Comparator, Outcome, Study design), and adds two practical checks: timing and certainty. The glossary defines PICO-S and GRADE if the terms are new.
Question 1: Who was studied, and how was the condition defined?
Population and definition come first because everything else depends on them. Lipoedema offers a clear example. The 2024 S2k guideline (Faerber and colleagues) describes clinical diagnosis and notes the absence of a laboratory biomarker, and the Field Guide records that criteria vary across research, expert statements, services and historical teaching. A study that requires one set of features and a study that requires another are not recruiting the same group, even if both use the word lipoedema.
The same applies to lymphoedema. The DiSipio systematic review (Lancet Oncology, 2013) concerns unilateral arm lymphoedema after breast cancer. A result from that population cannot be read as a result for lymphoedema in general.
Question 2: What was it compared with?
A comparator changes what a result means. A finding against "usual care" differs from a finding against another approach, and both differ from a before-and-after comparison with no control group. The Mortada systematic review of liposuction in lipoedema (Archives of Plastic Surgery, 2024) pools mostly observational cohorts and has no randomised comparison with conservative care. That is a design fact, not a verdict, and it limits what the pooled change can say about cause.
Question 3: What was measured, and with which instrument?
Volume, circumference, tissue measures, imaging and patient reports do not measure the same thing. The Hidding systematic review (Physical Therapy, 2016) looked at the measurement properties of instruments for lymphoedema and found heterogeneous evidence. Deltombe and colleagues (Lymphology, 2007) compared circumferential, water displacement and optoelectronic volumetry methods in a defined upper-limb population, which informs agreement between methods in that cohort and protocol.
Outcome choice matters just as much. A volume change and a change in pain or function are different outcomes. The Ezzo Cochrane review of manual lymphatic drainage (2015) found a modest volume signal in some defined settings and inconsistent findings for pain, function and quality of life. Papers that pick different primary outcomes can reach different headlines from similar data. The assessment chapter explains why reliability, validity and diagnosis are separate ideas.
Question 4: When was it measured, and for how long?
Timing can explain a surprising amount of disagreement. DiSipio and colleagues synthesised cohort studies in which incidence estimates varied with the definition used, the thresholds and instruments applied, the assessment points, the treatment groups and the length of follow-up. A figure taken soon after treatment answers a different timing question from one taken years later. A study that identifies change through a questionnaire may not classify it the way a study using limb measures does.
Follow-up also affects what is visible. Short follow-up cannot speak to long-term outcomes, and loss of participants over time changes who remains in the data. The Field Guide's management chapter treats attrition and follow-up as part of the result.
Question 5: What kind of study is it?
Study design sets the strongest claim a paper can make.
- Randomised trials and Cochrane syntheses compare defined groups and can speak to effects within their populations. The Ezzo review included six randomised trials, and the authors reported small, heterogeneous trials.
- Observational cohorts show change in selected groups but cannot rule out selection, co-interventions or other explanations.
- Cross-sectional studies describe a group at one moment. The Günay review of quality of life in lipoedema (Phlebology, 2025) reports evidence that is largely cross-sectional and often measured with generic instruments.
- Screening and survey studies reflect who responded and how the screen was built. Amato and colleagues (2022) used an online convenience screening sample in Brazil. It relied on self-report and is not a clinical diagnosis study or evidence for a UK figure.
- Consensus and guidelines organise expert judgement and literature. The International Society of Lymphology consensus (2020) and the S2k guideline are useful for terminology and clinical framing. They are not experimental validation.
- Qualitative studies describe what participants said. They do not estimate prevalence or outcomes.
When a media story or a clinic page reports "a study found", the design determines whether that sentence can carry the claim attached to it.
Question 6: How certain is the body of evidence?
Certainty looks across studies. Heterogeneity, inconsistency, imprecision from small samples, risk of bias and indirectness all reduce confidence. GRADE is a framework used to separate confidence in evidence from the strength of a recommendation. The point of using it, even informally, is to avoid reading a pooled number as more solid than the studies underneath it. A meta-analysis of weak or varied studies produces a tidy figure, and the tidiness is not evidence of strength.
A worked comparison, using no invented numbers
Picture two papers on swelling after breast-cancer treatment, one reporting a higher incidence than the other. Run the six questions. If the first measured with a questionnaire soon after treatment and the second used limb measurements over a longer follow-up, the difference may simply reflect definition and timing. If one included people with several treatments and the other only one, the populations differ. If one was a registry of a service caseload and the other a prospective cohort, the recruitment differs. After those checks, a gap that remains between comparable studies deserves real attention. A gap that dissolves has taught you about the studies, not about a conflict in the science. This scenario is illustrative and is not drawn from specific figures.
Where UK guidance fits in the framework
UK sources answer different questions from trials, so the framework applies to them in a different way. NHS pages on lymphoedema and lipoedema orient the public, so asking about comparators or follow-up misses their purpose. NICE guidance governs a named scope: NG101 for early and locally advanced breast cancer, NG141 for cellulitis and erysipelas antimicrobial prescribing, and HTG618 and HTG622 for liposuction in chronic lipoedema and chronic lymphoedema respectively. A NICE procedure document is a governance and evidence-safety position, not a review of every treatment. BLS and LSN materials add professional operational context and patient information. They are not systematic reviews. Keeping each source tied to its job avoids treating a public page as an effect estimate or a trial as a service rule. The UK services context page sets these roles out.
A short checklist to keep
- Who was in the study, and what definition admitted them?
- What was the comparison group, if any?
- Which outcome was primary, and how was it measured?
- When was it measured, and for how long were people followed?
- What design is this, and what is the strongest claim that design allows?
- How certain is the wider body of evidence, and who funded the work?
Changing or urgent concerns
Reading studies is not a substitute for care. A new, rapidly changing or worrying swelling or skin concern needs appropriate NHS or urgent-care advice, and fever, feeling unwell, breathing difficulty, chest pain or collapse needs emergency help.
The takeaway
Studies in this field differ because their populations, definitions, comparators, outcomes, timing and designs differ. Asking six questions turns an apparent argument into a clear list of reasons. What remains is the honest uncertainty, which the Field Guide names rather than hides, and the research index helps you follow.
Sources
- Faerber G et al. S2k guideline lipedema. J Dtsch Dermatol Ges. 2024;22(9):1303-1315. DOI 10.1111/ddg.15513.
- DiSipio T et al. Incidence of unilateral arm lymphoedema after breast cancer: a systematic review and meta-analysis. Lancet Oncology. 2013;14(6):500-515. PMID 23540561.
- Mortada H et al. Safety and effectiveness of liposuction modalities in managing lipedema. Archives of Plastic Surgery. 2024;51(5):510-526. DOI 10.1055/a-2334-9260.
- Hidding JT et al. Measurement properties of instruments for measuring of lymphedema: systematic review. Physical Therapy. 2016;96(12):1965-1981. DOI 10.2522/ptj.20150412.
- Deltombe T et al. Reliability and limits of agreement of circumferential, water displacement, and optoelectronic volumetry in the measurement of upper limb lymphedema. Lymphology. 2007;40:26-34. PMID 17539459.
- Ezzo J et al. Manual lymphatic drainage for lymphedema following breast cancer treatment. Cochrane Database Syst Rev. 2015;(5):CD003475. DOI 10.1002/14651858.CD003475.pub2.
- Günay UB, Talak GR, Demiröz A. Health-related quality of life among lipedema patients: a systematic review and meta-analysis. Phlebology. 2025. PMID 41436421.
- Amato ACM et al. Lipedema prevalence and risk factors in Brazil. J Vasc Bras. 2022;21:e20210198. PMC9136687.
- International Society of Lymphology. The diagnosis and treatment of peripheral lymphedema: 2020 Consensus Document. Lymphology. 2020;53(1):3-19. DOI 10.2458/lymph.4649.
- NICE. NG101, NG141, HTG618 and HTG622.
