Stage 1 Registered Report: Effects of high-intensity interval training in women with lipedema -  protocol for a randomized controlled trial with sequential intervention and control periods (The LipidEx study)

Primary aim

To determine whether a 12-week high-intensity interval training (HIIT) intervention reduces pain in women with lipedema compared to a control period of usual daily activities. Pain will be assessed using RAND-36 pain-related domains (primary outcome measure) and the Brief Pain Inventory (secondary outcome measure).

Secondary aims

To determine whether a 12-week HIIT intervention:

Hypotheses

The study is designed as a superiority trial.

For the primary aim, we hypothesize that a 12-week HIIT intervention will reduce pain more than usual daily activity.

For the secondary aims, we hypothesize that HIIT will

Goals and deliverables

The short-term goal is to generate initial evidence on the effects of HIIT in women with lipedema. The long-term goal is to inform and improve treatment guidelines for lipedema.

The main deliverables from the LipidEx include improved understanding of how HIIT affects pain, health-related quality of life, and adipose tissue in women with lipedema. The findings may have clinical relevance and the potential to inform future practice. By investigating both physiological and psychological outcomes, this project aims to provide a foundation for evidence-based, patient-centered treatment approaches.

Study approach and user engagement

This study follows the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) reporting guidelines.⁵¹ The user perspective in LipidEx is provided by the Norwegian Lymphedema and Lipedema Foundation (NLLF) and The Norwegian Research Centre for Women’s Health. Representatives from these organizations, selected for their long-term experience in women’s health and engagement with end-users, have been co-creators of LipidEx. They have been involved in the study design and will contribute throughout all phases, including recruitment, reporting of findings, and evaluation. These collaborators also participated in our previous feasibility study (ClinicalTrials.gov identifier: NCT05488977), providing valuable patient perspectives. Their involvement aims to optimize study relevance and feasibility, enhance protocol quality, and potentially reduce dropout rates.

Study design and participants

A preprint version of the study protocol has previously been published on medRxiv.⁵² We will conduct a randomized controlled trial (RCT) with sequential intervention and control periods with 58 women diagnosed with lipedema. To be eligible for participation, women must be between 18 and 65 years old and have a confirmed diagnosis of lipedema established by qualified healthcare personnel prior to study inclusion. Participants with all stages and types of lipedema will be eligible for inclusion. Exclusion criteria include self-reported ongoing eating disorders, pregnancy, regular endurance training, current medications for weight loss, and/or orthopedic limitations for exercise training. Participants will be allowed to continue their usual medical care during the trial. However, they will be excluded if they undergo surgery for lipedema during the study period. We have defined the following terms: I) Ongoing eating disorder - a condition that has been diagnosed, or should be assessed, by a healthcare professional, which causes significant daily challenges related to food, body, or health. Examples include anorexia, bulimia, binge-eating disorder, or other serious and persistent eating-related issues, and II) Regular endurance training - activities that elevate the heart rate, such as brisk walking, jogging, cycling, swimming, or similar exercises, performed at least twice a week per week over the past four weeks.

The participants will be randomized (1:1) using restricted block randomization to either Training Group 1 or Training Group 2. Training Group 1 will start with 12 weeks of HIIT, tailored to each participant’s fitness level and the severity of their lipedema, while Training Group 2 will continue their usual daily activities. Following the 12-week intervention, Training Group 1 will enter a 12-week interim period (a so-called washout period) before proceeding to a 12-week control period. During both the washout and control periods, participants will have no study-related contact and will not take part in any organized exercise sessions provided by the study. They will be asked to maintain their normal daily routines, and no specific instructions or restrictions regarding physical activity will be given. Training Group 2 will begin their 12-week HIIT intervention directly after completing their control period. Each participant will attend 3 or 4 test days, depending on group allocation, to complete the study-related measurements and testing (Supplementary Figure 1). The total study duration will be approximately 24 or 36 weeks, depending on the assigned group, with study initiation from September 1^st, 2025, and expected completion by August 2026.

Due to the nature of the intervention, neither participants nor investigators can be blinded to group allocation. However, several procedures are in place to minimize bias. Randomization will be conducted using restricted block randomization with a fixed initial block (larger than the subsequent ones) followed by varying block sizes (two different sizes). The randomization sequence was generated by an independent third party not involved in participant enrollment or data collection, and implemented using the secure online platform eFORSK, which ensures allocation concealment until completion of baseline assessments. No stratification will be applied. Study personnel responsible for enrollment and testing will not have access to the allocation sequence before randomization, and participants will be informed of their group assignment only after completing their first test day. Outcome assessors and data analysts will remain blinded to group allocation throughout data analysis. Standardized protocols will be used for all testing procedures and outcome assessment to further minimize potential bias.

Recruitment

Participants will be recruited in Trøndelag, Norway, primarily through The Clinic of Rehabilitation at St. Olavs Hospital and NLLF. Recruitment posters will be displayed at general practitioner offices and various physiotherapy clinics in the Trondheim area and spread thorough the social media of NLLF and the internationally renowned Cardiac Exercise Research Group (CERG). Interested participants can register via a QR code on the recruitment poster and will be contacted shortly thereafter to determine eligibility. Subsequently, they will receive detailed information about the study via e-mail to evaluate their interest in participation prior to study start.

Exercise intervention

Exercise testing and training will be conducted at state-of-the-art exercise training facility of NTNU at St. Olavs hospital, the NextMove Core Facility, supported by highly experienced staff. Participants will perform three HIIT sessions per week for 12 weeks. HIIT session will consist of four 4-minute intervals at 85–95% of maximal heart rate (HR_max), separated by 3-minute active recovery periods (~60% HR_max). Two of the weekly HIIT sessions will be supervised at the NextMove Core Facility and primarily performed on treadmills. Spinning bikes will be available when individual adaptations are needed due to pain, discomfort, or functional limitations. The third weekly session each week will be unsupervised and standardized for intensity and duration (16 minutes of high-intensity exercise, equivalent to four 4-minute intervals), allowing participants to choose a preferred exercise modality such as uphill walking, cycling, swimming, rowing, or similar activities. Swimming will be specifically recommended due to its potential benefits related to natural compression and reduced joint loading In the intervention period, activity will be monitored using Polar belt (Polar, Polar Electro, Kempele, Finland) in the supervised sessions and Xiaomi Smart Band 10 heart rate monitor in the unsupervised sessions. The participants will retain Xiaomi Smart Band 10 after study completion to promote continued physical activity.

Individualized training sheets will be developed based on each participant’s HR_max and VO_2max/peak obtained from the test day before their intervention period, specifying the target heart rate zones for warm-up, work intervals, and recovery periods. During supervised sessions, personnel will record heart rate data in a standardized template, noting the heart rate during the final minute of each interval and recovery period to ensure compliance with the prescribed training intensity (85–95% HR_max during intervals and ~ 60% HR_max during recovery). These data will also be used to adjust treadmill/spinning bike speed and/or incline/resistance to maintain the correct workload. For unsupervised sessions, compliance will be assessed through post-session discussions with participants and potential review of heart rate data from the Xiaomi Smart Band 10, ensuring that participants have performed the exercise and reached their prescribed heart rate zones. Adherence to the exercise intervention will be considered adequate if ≥80% of sessions are completed. While compliance with the prescribed intensity zone of 85–95% of HR_max is not required for adequate adherence, our monitoring procedures and follow-up will ensure that participants generally achieve the intended intensity during training sessions.

Participants will receive an exercise diary to record details of all unsupervised sessions throughout the intervention period. Additionally, they will complete a standardized template provided by NextMove Core Facility on weekly basis, documenting whether the session was completed, the total duration of high-intensity intervals, whether the target heart rate zone (85–95% of HR_max) was achieved, and any additional notes in case of uncertainty. Study personnel will review the completed templates on-site, and when possible, verify participants’ session data through the activity monitor app. This standardized template ensures consistent reporting across participants and allows for verification of compliance during the intervention. During the washout and control periods, participants will not maintain exercise logs but will be asked to report their general activity patterns on the following test days.

Data collection

During the first study visit (Test day 1), participants will be randomized into Training Group 1 or Training Group 2 and receive comprehensive information about the study and all included procedures. Data collection will occur primarily on three or four separate test days, depending on the allocated Training Group, with a duration of approximately 3 hours. On these test days, anthropometric and physiological measurements (blood pressure and VO_2max) will be taken, questionnaires will be completed, and fasting blood samples will be collected. Adipose tissue samples will be collected for those who consent, on the test day before and after the intervention period (two samples collected per participant in total). General information, including details on lipedema (type and morphological stage, and time of diagnosis), comorbidities, and the use of medications, supplements, and clinical aids (e.g., compression aids and their frequency of use), will also be gathered on all test days. Information regarding lipedema will be based on the participant’s medical assessment and diagnosis established by qualified healthcare personnel prior to study participation. During the intervention period, the participants will be asked to fill out a short questionnaire at their last training session at St. Olavs hospital each week to gather information about perceived pain and health-related quality of life. All data from the test days will be registered on eFORSK.

Pain registration

Pain, the primary endpoint, will be addressed at all test days using two validated instruments: RAND-36 health survey (Pain Dimension)⁵³ and the Brief Pain Inventory (BPI).⁴⁶ RAND-36 will serve as the primary outcome measure for pain, while BPI will serve as the secondary outcome measure.

RAND-36 (also known as SF-36) includes two questions designed to measure the intensity of pain and the extent to which pain interferes with normal work (both outside the home and housework). The first question assesses the intensity of pain experienced over the past four weeks with six response options from “none” to “very severe”. The second question evaluates the degree to which pain has interfered with daily activities and work with five response options from “not at all” to “extremely”.

BPI includes four questions that assess the severity of pain right now and at its worst, least, and average. Each question gives a possible answer of 0 to 10, where 0 means no pain and 10 is the worst pain you can imagine. The four BPI questions contribute with the same weight, and they are combined to a final pain severity score with a range from 0 to 40. The results from RAND-36 and BPI will be evaluated separately. Whereas both tools assess pain, the BPI provides a deeper and more detailed assessment of pain characteristics and functional impact compared to the broader, more generalized assessment provided by the RAND-36 pain dimension. However, the questionnaires focus on different timeframes, as RAND-36 assesses pain over a retrospective period of the past 4 weeks whereas BPI assess pain over the past 24 hours or at the present moment (current pain intensity). Potentially, RAND-36 can be more useful for assessing chronic pain, and BPI for assessing more current pain intensity. However, the most suitable tool for assessing pain specifically in lipedema has yet to be determined.

Anthropometric measurements

Body composition, including weight (kg), body mass index (kg/m²), muscle mass (kg), total adipose tissue mass (%), adipose tissue mass in the lower limbs (%), basal metabolic rate (kcal), and mineral content (kg), will be measured non-invasively using the bioelectrical impedance instrument InBody 770 (Biospace CO, Ltd, Seul, Korea). All assessments will be conducted in the morning (before 11:00 a.m.) according to a standardized protocol. Participants will arrive in a fasted state, void their bladder before measurement, and wear light clothing. They will stand barefoot on the device and follow the manufacturer’s on-screen instructions. Participant ID, sex, age, and height will be entered prior to the scan, and height (cm) will be measured without shoes before the bioelectrical impedance analysis.

The circumference (cm) of the waist, hip, right thigh, and right calf will be measured using standardized procedures at NextMove Core Facility. The waist-to-hip ratio (cm/cm) will then be calculated. A measuring board will be used to ensure consistent measurement sites for the thigh, calf, hip, and waist at all test days.

Quality of life

The RAND-36 questionnaire will be used to detect potential changes in health-related quality of life and data will be collected at all test days.⁵³ This questionnaire includes 36 questions covering eight health domains, including vitality, physical functioning, bodily pain, general health perceptions, physical role functioning, emotional role functioning, social role functioning, and mental health. Each domain is scored on a scale from 0 to 100, where the scaled scores are weighted sums of all questions within each of the eight domains. A higher score indicates better health-related quality of life.

Physiological measurements

Blood pressure will be measured in a sitting position, after some minutes rest in a quiet room, with a digital blood pressure monitor (Welch Allyn ProBP 2400, Helsinki, Finland) by trained personnel. Two cuffs’ sizes were used, size 11 (25–34 cm) and size 12 (32–43 cm). The blood pressure will be measured at the same time-point of the day for each participant at all test days. The digital blood pressure monitor performs three readings. The first reading will be discarded, and the mean of the two last readings will be used.

Cardiopulmonary exercise testing (CPET) (Metalyzer ll, Cortex Biophysik GmBH, Leipzig, Germany) will be performed on a treadmill (Woodway treadmill 3, Weil am Rhein, Germany) to measure VO_2max/peak under supervision by qualified personnel. An individualized protocol will be used, starting with a 10-minute warm-up (~60% of HR_max) to determine the appropriate workload, followed by increased workload every minute until exhaustion. The workload will be increased by either 1 km/h in speed or 2% incline per minute. The choice of whether to increase speed or incline will be determined by qualified personnel based on the participant’s warm-up, exercise history and tolerance, and orthopedic limitations. Borg Rated Perceived Exertion scale, a subjective measure scale of perceived exertion from 6–20, will be recorded during warm-up and immediately after the CPET.³¹ Capillary blood lactate concentration will be measured (Biosen C_line, EKF diagnostic GmbH, Barleben, Germany) before warm-up, after warm-up, and at maximal exhaustion. Heart rate will be monitored continuously using a Polar heart rate belt and watch (Polar, Polar Electro, Kempele, Finland). The highest measured heart rate will be used to determine HR_max. Criteria for reaching VO_2max is a levelling off of oxygen uptake (VO₂) despite increased workload and a respiratory exchange ratio ≥ 1.05. Blood lactate concentration will be used as a supplementary indicator of maximal effort, with ≥8 mmol/L serving as a supporting criterion.⁵⁴

Blood samples and adipose tissue biopsies

Five venous blood samples will be collected in a fasting state (minimum 8 hours) by qualified personnel following standard in-hospital procedures at St. Olavs hospital. Directly after collection, two 3.5 mL EDTA-tubes and one 6 mL Li-heparin-tube (serum) will be sent to routine biochemical analyses at St. Olavs hospital for measuring glucose, glycated hemoglobin, triglycerides, high-density lipoprotein (HDL) cholesterol, total cholesterol, low-density lipoprotein cholesterol, non-HDL cholesterol, lipoprotein (a), thyroid-stimulating hormone, free thyroxine (FT4) and high-sensitivity C-reactive protein . The two remaining tubes, one 6 mL serum and one 6 mL EDTA tube, will be centrifuged at 2200 g/rpm for 10 minutes at 20 °C, before being further aliquoted and stored in the LipidEx biobank at −80 °C for future explorative analyses. EDTA will be centrifuged immediately after collection, while serum will be centrifuged 30–60 minutes after collection.

Two adipose tissue samples will be collected by experienced personnel from the Research outpatient clinic at St. Olavs hospital before and after the exercise intervention. The collected adipose samples will be used to advance our physiological understanding of exercise-induced effects on adipose tissue in women with lipedema. To ensure patient comfort during the procedure, local anesthesia will be administered at the site of biopsy to minimize pain and discomfort associated with the needle insertion. The proposed methodology will adhere to rigorous sterilization protocols, including the use of sterile equipment, gloves, and disinfection procedures. A fine needle is used to avoid the need for extensive incisions and to reduce patient discomfort and recovery time. The adipose tissue sample will be divided into two parts: One will be snap-frozen in liquid nitrogen and stored at −80 °C for later explorative analyses, and the other will be placed in histology cassettes and fixed in 4% formaldehyde in 0.1 M phosphate buffer for 48 hours. The samples will be dehydrated, embedded in paraffin, sectioned, and stained with hematoxylin and eosin in the Histology laboratory of the Cellular & Molecular Imaging Core Facility (CMIC) at the Department of Clinical and Molecular Medicine (IKOM), NTNU for later explorative analyses.

Adverse events and safety monitoring

A risk assessment and management plan has been developed in collaboration with the medically responsible physician for LipidEx. Participation in LipidEx is considered to involve minimal risk.

All adverse events will be systematically recorded throughout the study period. Participants will be informed at study start and reminded prior to each supervised session to report any discomfort, injury, or unexpected reactions occurring during exercise testing, training, or adipose tissue sampling. Personnel and the medically responsible investigator will evaluate reported adverse events with respect to severity, causality (related/unrelated to the intervention), and outcome. Any serious adverse events will be reported immediately to the principal investigator, who will, if required, notify REK. All events will be documented in the electronic case report form (eCRF) within the eFORSK system, and serious incidents will be reported to the research leadership within 24 hours of awareness. A summary of all recorded adverse events, including their frequency, severity, and relationship to the study intervention, will be reported in the final publication.

Given the relatively short duration and low-risk nature of the intervention, no formal Data Safety Monitoring Board or predefined stopping guidelines have been established. The research team will continuously monitor safety and feasibility, and the principal investigator will decide on any necessary trial modifications or discontinuation in consultation with REK.

All supervised training sessions are conducted at the NextMove Core Facility at St. Olavs hospital, NTNU, under the supervision of experienced personnel trained in clinical exercise testing and emergency procedures. Safety precautions include continuous heart rate monitoring and symptom surveillance. A separate risk assessment plan has been developed for adipose tissue sampling to minimize the risk of infection, bleeding, or discomfort.

Sampling plan

Statistical power

As lipedema is a disease that just recently has received more attention in research, there is limited data available in the literature to make an accurate prediction of the expected size of the effect of HIIT on pain levels in women with lipedema. To identify input values for sample size calculation, we used preliminary RAND-36 pain-domain data from our 8-week feasibility study (unpublished trial, ClinicalTrials.gov: NCT05488977, 2022–2023). Previous reports suggests that a minimal clinically important difference for RAND-36 domains typically range from 5 to 10 points.⁵⁵ The mean self-reported pain level at baseline for the lipedema patients in our feasibility study was 55 points (SD = 23), which were similar to previously reported pain levels for women with lipedema.^11,56,57 Following the 8-week intervention period in the feasibility study, the intervention group improved to 69 points, corresponding to a 14-point change (inversed pain scale). Based on this observed difference, we considered a minimum of 10-point improvement in the RAND-36 pain domain to be clinically relevant for lipedema patients in the present study. Assuming a standard deviation of 23 and a within-subject correlation of 0.5, a total of 44 participants would provide 80% power at a 0.05 significance level*. Accounting for a potential dropout rate of 30%, we plan to include 58 participants. Calculation was performed using Stata version 18 (StataCorp, College Station, TX, USA).

* The calculation is made for assuming a paired sample t-test, but the analysis will be performed within the framework of a linear mixed model (LMM).

Sensitivity analyses for the BPI will be performed using preliminary data from the feasibility study, alongside published values for minimal clinically important difference, to assess the clinical relevance of observed changes.

Statistical analysis plan

Pre-processing steps

Data is collected by the electronic case report form (eCRF) eFORSK. After data cleaning (searching for data entry errors or outliers), the database will be locked, and data will be extracted as sav files and csv files for SPSS and R, respectively. The following statistical analyses will be performed using SPSS v.21 and RStudio. All findings will be presented using a significance level of 0.05, and p-values adjusted for multiple testing will also be reported to account for multiplicity.

Efficacy of the intervention will be assessed according to the intention-to-treat principle, in which participants will be analyzed in the group to which they were originally randomized, regardless of adherence to the intervention. In addition, per-protocol analyses will be conducted including participants who have completed at least 80% of the planned exercise training sessions (both supervised and unsupervised). No data imputation is planned for missing data; analyses will be based on available data. The extent and pattern of missing data will be reported, and sensitivity analyses may be conducted where appropriate. Interim analyses are not planned. Sample characteristics will be summarized using frequencies and percentages, medians with interquartile range, or means with standard deviation, as appropriate.

Planned analyses

Primary endpoint

We will apply an LMM with repeated measures (random intercept model) to compare the changes in pain score during the intervention period compared to the changes within the control period. Model assumptions will be assessed by using residual diagnostics. If model assumptions are found not to be met, LMMs on transformed data or alternative methods will be considered.

Secondary endpoints

For the three secondary endpoints, adipose tissue mass (both total and in the lower limbs) and health-related quality of life, similar LMMs as described for the primary endpoint will be applied.

Additional analysis

Adipose tissue mass impact on pain score

If a significant intervention effect on pain is observed, we will perform mediation analysis to explore whether changes in secondary outcomes related to adipose tissue mass (total adipocyte mass (%) and adipocyte mass in lower limbs (%)) mediate the relationship between HIIT and change in pain score.

Effect of intervention order

To investigate whether the randomized order of the intervention modified the intervention effect, an exploratory analysis will be carried out by extending the LMM for the primary outcome to include an interaction with order.

Limitations

This study has several methodological and practical limitations that should be acknowledged. The sample size is based on preliminary feasibility data and is powered to detect changes in pain levels within the intervention group. Therefore, it may be underpowered to detect smaller between-group differences in secondary outcomes. As recruitment is limited to women with lipedema from a single site in Norway, the external validity and generalizability of the findings may be restricted. Blinding of participants and study personnel is not feasible due to the nature of the intervention, which may introduce performance and expectation bias, particularly for self-reported outcomes such as pain and health-related quality of life. Adherence to the unsupervised training sessions may vary, and although close follow-up will minimize this risk, individual differences in compliance and exercise execution could influence the results. Finally, the 12-week intervention period captures short-term effects only, and the sustainability of potential benefits beyond this period remains unknown.

Dissemination plan

The target audience, stakeholders, and end-users of LipidEx incudes women, lipedema patients, healthcare professionals, researchers, the pharmaceutical industry, insurance companies, regulatory authorities, policy makers, and other public bodies. The NLLF, represents the societal burden of lipedema among women in Norway, and they will act as the main audience, stakeholder groups and potential end-users of the research results. These partners have a large network of patients and healthcare professionals, as well as a long-standing relationship with policy makers, and will therefore be crucial in disseminating the results through their appropriate channels (e.g., webpages, brochures, events). We also aim to promote the results internationally through the channels of our international collaborators at the Foeldi clinic and International Lipoedema Association. In addition, our research group has a long tradition for outreach activity and has a dedicated communications adviser who will contribute to promote LipidEx both nationally and internationally. Several targeted communication and dissemination measures will be applied by using appropriate channels such as X, Instagram, webpages, Open Access publication and presentations at both national and international conferences. Together, these efforts will ensure efficient and sustainable dissemination of the generated knowledge throughout the community. LipidEx will ensure open science, and strictly follow the FAIR principles (Findable, Accessible, Interoperable and Re-usable).

Protocol amendments and trial termination

Any modifications to the study protocol that may affect the conduct of the trial, participant safety, or scientific validity (e.g., changes in study objectives, design, eligibility criteria, or analysis plan) will be formally documented and submitted for approval to the REK and updated in the trial registry and preprint server. All relevant stakeholders, including participants, will be informed of substantial changes as appropriate. The study may be terminated prematurely by the investigators in the event of unforeseen safety concerns or lack of funding. In such cases, procedures will be followed to ensure participant safety, proper data management, and transparent reporting of the reasons for termination.