Article Text

Download PDFPDF

Carbon dioxide laser excision as a novel treatment for large xanthelasma palpebrarum: long-term efficacy and safety
  1. Dingqiao Wang1,
  2. Zhen Mao1,
  3. Zuohong Li1,
  4. Wuyou Gao1,
  5. Yixin Qu1,
  6. Xingyi Li1,
  7. Yuzhen Jiang1,2,
  8. Xianchai Lin1
  1. 1State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, Guangdong Provincial Key Laboratory of Ophthalmology Visual Science, Guangdong Provincial Clinical Research Center for Ocular Diseases, Guangzhou, China
  2. 2Ophthalmology, National Institute for Health Research Biomedical Research Centre at Moorfields Eye Hospital NHS Foundation Trust, UCL Institute of Ophthalmology, London, UK
  1. Correspondence to Professor Xianchai Lin; linxch7{at}mail.sysu.edu.cn; Dr Yuzhen Jiang; sanny{at}aimirainnovation.com; Dr Xingyi Li; lixy237{at}mail2.sysu.edu.cn

Abstract

Aims This study evaluates the long-term efficacy and safety of carbon dioxide (CO₂) laser excision as a novel treatment for large xanthelasma palpebrarum.

Methods The study included 295 patients diagnosed with xanthelasma palpebrarum, categorised as 246 patients with grade I, 16 with grade II, 22 patients with grade III and 11 with grade IV lesions. All underwent CO₂ laser excision and were followed up for a 12-month period. Treatment efficacy was assessed through clearance and recurrence rates, evaluated using digital photography and complications were recorded.

Results The participant group consisted of 66 males and 229 females, with an average age of 41.7±11.6 years. Clearance rates exceeded 99% for grades I–III and were around 95% for grade IV lesions. The recurrence rate was 6.8%, with higher recurrence rates observed in lesions over 2 mm in height compared with those under 2 mm (p<0.001). The main complications over the 12-month follow-up were scarring (4.4%), hyperpigmentation (8.1%) and hypopigmentation (8.5%), with no severe complications reported.

Conclusion CO₂ laser excision emerges as a precise, minimally invasive and effective modality for treating xanthelasma palpebrarum, marking a significant advancement in treatment modalities. Further research is required to reinforce these results and refine treatment protocols.

  • Treatment Lasers
  • Eye Lids
  • Ophthalmologic Surgical Procedures

Data availability statement

Data are available on reasonable request. The data including deidentified participant photographs are available on reasonable request (wangdq5@mail2.sysu.edu.cn).

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

WHAT IS ALREADY KNOWN ON THIS TOPIC

  • Before our study, laser ablation was commonly employed for treating xanthelasma palpebrarum, however, it had shortcomings, especially when dealing with extensive grade IV lesions. Although surgical excision was effective for larger lesions, it frequently necessitated complex reconstructive procedures, which increased the risk of complications and prolonged recovery time.

WHAT THIS STUDY ADDS

  • Our study introduces carbon dioxide (CO2) laser excision as an innovative approach for xanthelasma palpebrarum, demonstrating particular efficacy for large, grade IV lesions. The study reveals a high clearance rate exceeding 95%, along with a low recurrence rate of 6.8% and minimal complications, such as scarring (4.4%), hyperpigmentation (8.1%) and hypopigmentation (8.5%).

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

  • The CO2 laser excision technique, which employs the laser for precise cutting, presents a substantial improvement over traditional methods.

Introduction

Xanthelasma palpebrarum, a benign lesion commonly found on the eyelids, presents as yellowish, flat plaques, often located near the inner canthus.1 This condition, which is the most frequently occurring type of xanthoma on the skin, affects nearly 1.1% of women and 0.3% of men.2 ,3 While it poses no medical threat, the lesions can grow and become noticeable, leading many individuals to seek treatment for cosmetic reasons.

A variety of approaches have been employed to treat xanthelasma palpebrarum, ranging from traditional surgical removal to modern techniques such as laser therapy, electrosurgery, chemical peeling, cryotherapy and targeted injections.1 4 5 A systematic review of xanthelasma palpebrarum, encompassing 49 studies, indicates a preference for retrospective analysis and single-arm study designs.6 For extensive xanthelasma, a combination of surgical removal, eyelid reconstruction and skin grafting is often suggested.7 Among laser treatments, the carbon dioxide (CO₂) and Er:YAG lasers have shown promising results, with a significant percentage of patients experiencing substantial improvement.8 The CO₂ laser tends to outperform the Er:YAG laser and trichloroacetic acid peeling, although it may cause potential complications such as changes in skin pigmentation.9 ,10

Previous studies have indicated that CO₂ laser ablation is a commonly used treatment method (online supplemental table 1).4 11–14 The challenge arises from the fact that lower laser energies result in smoother treatment surfaces but require longer treatment times of around 20–30 min. Conversely, higher laser energies reduce vaporisation time but can lead to unevenness in the treated area. Consequently, laser ablation is primarily used for smaller lesions within grade III.6 15 Especially for lesions deeper into the dermis, recurrence was a concern with laser ablation.5

Supplementary video

To address these limitations, this study explores the use of CO₂ laser excision for xanthelasma palpebrarum, a method that cuts and removes the lesion entirely from its base rather than vaporising it layer by layer.16 17 This approach aims to provide a more efficient and effective treatment for large lesions, with improved outcomes and reduced treatment times. Our retrospective analysis of 295 patients treated with this method assesses its efficacy and safety, particularly for larger xanthelasma lesions, over a 12-month follow-up period.

Materials and Methods

Study design and participants

We conducted a retrospective analysis at the Zhongshan Ophthalmic Center, reviewing 295 cases of xanthelasma palpebrarum diagnosed from June 2013 to June 2023. Patients sought treatment primarily for aesthetic reasons or due to discomfort. Informed consent for CO₂ laser therapy and photographic documentation was obtained from each patient before treatment . Follow-up evaluations were scheduled at 1 month and 12 months post-treatment.

Inclusion criteria involved a clinical diagnosis of xanthelasma palpebrarum, absence of photosensitivity, bleeding, or coagulation disorders, and completion of at least 12 months of follow-up. Exclusion criteria included any pre-existing scarring or infection at the intended treatment site, lidocaine allergy, presence of malignant lesions and systemic immunological diseases.

Demographic data such as age, gender, medical history, medication use and any familial history of xanthelasma palpebrarum and hyperlipidaemia were recorded. Lipid profiles were measured in the fasting state. The clinical assessments categorised the xanthelasma lesions (number, size, site and extent) and established the grading system as follows15: grade I designates lesions confined to the upper palpebral region; grade II extends to the medial canthus; grade III presents on the medial aspects of both upper and lower palpebral regions and grade IV signifies lesions presenting on both medial and lateral aspects of the upper and lower palpebral regions.

Lesion sizes were evaluated by using a calliper to measure the length and width at its longest point while holding the skin taut. Additionally, the height of the lesion was documented by the mirror device introduced by Nagamatsu et al.18 This method involves placing a device with a 45° angled mirror beside the base of the lesion and reading the height scale on the tilted mirror surface. For cases with multiple lesions, the maximum height among all lesions was recorded. The number of laser treatments was recorded. Treatment frequency varied, with some patients undergoing multiple sessions, ranging from two to more than four, depending on the individual case requirements.

Laser excision technique

Equipment and preoperative preparation

The study used a 10 600 nm CO₂ laser system (Ultrapulse, equipped with a DeepFx handpiece; Lumenis, Santa Clara, California, USA) for all procedures. Prior to the excision, local anaesthesia was administered using 1% lidocaine with epinephrine at a dilution of 1:100 000. In instances where the lesion was proximal to or involved the eyelid margin, lidocaine eye-drops were applied to anaesthetise the ocular surface.

Laser excision procedure

Pulse stacking was performed during the laser excision process. The procedure began with precise demarcation of the treatment area by vertically ablating 1 mm around the lesion’s perimeter. We used a continuous pulsed mode with each pulse having a duration of 10 ms. The laser energy was set between 100 and 200 mJ and adjusted based on the tissue response. Regarding the laser beam angle, it was nearly 45° when making the initial incision to create a flap. After the incision, the laser was used to separate the lesion from its base at an angle of less than 30° to nearly parallel to the skin surface. The laser spot size was generally 0.2 mm but was slightly adjusted within a small range depending on the size and thickness of the lesion. The objective was to remove the lesion effectively while avoiding damage to the dermis or the superficial layer of the orbicularis muscle, thus minimising the risk of scarring. Figure 1 provides a schematic of the laser excision technique, and a surgical video is available in online supplemental file 2.

Figure 1

Laser excision technique for xanthelasma palpebrarum. The left panel offers a schematic illustration, highlighting the precise removal process of the xanthelasma lesion from the underlying dermal layers. The right panel provides an actual intraoperative photograph. The laser beam angle is around 45° when making the initial incision to create a flap. After the incision, the laser is used to separate the lesion from its base at an angle of less than 30° to nearly parallel to the skin surface. It demonstrates the advantages of laser excision over traditional surgical methods, particularly in terms of achieving a smoother postoperative surface and thus reducing the risk of scarring. CO2, carbon dioxide.

The excision size was limited to 10×10 mm2 per eye to minimise the risk of difficulty in wound healing and scarring. The number of treatment sessions was determined by the size of the xanthelasma lesions, with an interval of at least 1 month between sessions.

Postoperatively, patients were instructed to keep the wound clean and dry. Local antibiotic ointment (0.3% tobramycin) was applied twice a day for 7–14 days. The wounds were observed to be epithelialised after 1–2 weeks.

Comparative analysis

We delineate the distinctions between laser excision, traditional laser ablation and surgical excision. Unlike the layer-by-layer vaporisation characteristic of conventional laser ablation, our laser excision technique employs a direct incision at the lesion’s base. This approach not only differentiates abnormal from normal tissue but also leverages the CO₂ laser’s haemostatic properties to achieve a clear, bloodless field. This method is similar to surgical excision but offers distinct advantages in precision and recovery. The schematic of the laser excision technique is visually represented in figure 1 for a comprehensive understanding.

Outcome assessments

The primary measures for evaluating the effectiveness of the CO₂ laser excision in this study were the clearance rate of the xanthelasma lesions and the rate of recurrence. To ensure objectivity, two independent plastic surgeons assessed these outcomes using a series of photographs taken before and after treatment.

Standardised photographs were taken by a trained technician using a consistent digital camera setup (Canon EOS 6D Mark II, Canon, Tokyo, Japan). The camera settings were kept uniform across all photographs. Participants were positioned at a fixed distance of 1 m from the camera, with consistent background and lighting conditions. Each participant was photographed from three angles: front, right side and left side.

Clearance rate evaluation

The extent of lesion clearance was quantified and categorised into four distinct levels: less than 50% cleared, 50%–75% cleared, 76%–95% cleared or greater than 95% cleared. This classification was determined by analysing the preoperative photographs against those taken after the final treatment session, calculating the proportion of the lesion that had been successfully removed relative to its total surface area.

Recurrence rate determination

Recurrence was specifically defined for this study as the reappearance of xanthelasma within the area that had been previously treated. The emergence of new lesions outside of the treated zones was not classified as a recurrence. This distinction was crucial for accurately measuring the success of the excision procedure.

Complication monitoring

Complications were monitored over a 12-month period following the procedure. Two ophthalmologists independently reviewed postoperative photographs to identify any complications, which included but were not limited to recurrence, ectropion, lagophthalmos, hypertrophic scarring and pigmentary changes. In cases of discrepancy between the two reviewing ophthalmologists, a senior ophthalmologist was consulted to make the final determination.

Documentation of treatment sessions

Additionally, the study documented the number of treatment sessions each patient underwent to achieve the desired outcome.

Statistical analysis

Data were processed by using SPSS V.23.0. Categorical variables were presented as percentages, and the χ2 test was used for comparative analysis. Multivariable logistic regression analyses were performed to assess the relationship between lesion parameters at baseline and risk of lesion clearance. A p<0.05 was deemed statistically significant.

Results

Efficacy outcomes

Our cohort consisted of 295 individuals diagnosed with xanthelasma palpebrarum, comprising 66 males and 229 females. The average age was 41.7±11.6 years, with a range from 26 to 77 years. The distribution of xanthelasma grades among the patients was as follows: 246 with grade I, 16 with grade II, 22 with grade III and 11 with grade IV lesions. Bilateral lesions were present in 185 patients while 42 had lesions exclusively on the right eye and 68 on the left eye. A total of 35 patients presented with lesions on both upper and lower eyelids, 254 had lesions only on the upper eyelid and 6 on the lower eyelid. Lesion heights were less than 2 mm in 278 patients and greater than 2 mm in 17 patients. Treatment-naïve patients numbered 277, and 18 sought treatment for recurrent lesions. The demographic and lesion data are detailed in table 1.

Table 1

Demographic characteristics and clinical data of 295 patients with xanthelasma palpebrarum

The efficacy of the laser treatment was high, with lesions of grades I–III showing a clearance rate surpassing 99% post-treatment. Grade IV lesions, particularly those with a height exceeding 2 mm, had a slightly lower clearance rate of 95%. Recurrence was observed in 20 patients (6.8%) during the 12-month follow-up. The number of laser sessions required for complete lesion clearance ranged from 1 to 7, with 171 patients achieving clearance after a single session, 109 patients needing 2–3 sessions and 15 patients requiring more than 3 sessions. These treatment outcomes are summarised in table 2.

Table 2

Treatment outcomes of CO₂ laser resection among subgroups of xanthelasma patients

We performed a subgroup analysis based on lesion grading and the size of grade IV lesions. Recurrence rates were 6.5% for grade I, 6.3% for grade II and 9.1% for grade III. Grade IV lesions were divided into two groups based on the total lesion area across both eyes: seven cases had a total area under 10 cm² and four cases had a total area above 10 cm². In the smaller area group, one case (14.3%) experienced recurrence while no recurrences were noted in the larger area group. The difference in recurrence rates between these subgroups was not statistically significant (p=0.957). Regarding the number of treatments, most grade I lesions (63.4%) cleared after a single laser session while most grade II lesions (62.5%) required 2–3 sessions. The majority of grade III lesions (81.8%) were cleared under three sessions. For grade IV lesions under 10 cm², 71.4% required more than three sessions and all grade IV lesions over 10 cm² needed more than three sessions for complete clearance. A higher grade and larger lesion size were associated with an increased number of treatment sessions (p<0.001).

When analysing based on lesion height, we found that recurrence occurred in 14 cases (5.0%) with lesions under 2 mm in height and in 6 cases (35.3%) with lesions over 2 mm. The recurrence rate was significantly higher in the group with lesions over 2 mm (p<0.001). For lesions under 2 mm, only 3.2% of patients required more than three treatment sessions while 35.3% of patients with lesions over 2 mm needed more than three sessions for complete clearance, indicating that greater lesion height necessitated more treatment sessions (p<0.001).

Post-treatment complications

Following the CO₂ laser excision, patients typically experienced transient symptoms including redness, swelling and a slight burning sensation at the treatment site, which generally subsided within 3 days. Throughout the 12-month monitoring period, we identified three primary complications associated with the procedure: hypertrophic scarring, observed in 13 participants (4.4%); hyperpigmentation, noted in 24 individuals (8.1%) and hypopigmentation, affecting 25 patients (8.5%). Notably, no severe adverse events, such as infections or damage to the lacrimal punctum, were reported.

A χ2 analysis was conducted to explore the variation in complication rates across different subgroups (table 2). Further, we segmented the data by lesion grades for a nuanced understanding of complication risks. The incidence of scarring across the grades—grades I–IV (with distinctions based on lesion size)—showed a trend towards increased scarring with higher lesion grades, with rates ranging from 3.3% in grade I to 50.0% in grade IV lesions larger than 10 cm² (p=0.002). Hyperpigmentation rates varied from 7.7% in grade I to 14.3% in grade IV (size ≤10 cm²), though these differences did not reach statistical significance (p=0.918). Similarly, the distribution of hypopigmentation across grades did not demonstrate significant variability (p=0.419).

When analysing complications by lesion height, significant differences were observed. For lesions under 2 mm in height, the rates of scarring, hyperpigmentation and hypopigmentation were 2.9%, 7.2% and 7.6%, respectively. In contrast, lesions exceeding 2 mm in height exhibited significantly higher rates of these complications: 29.4% for scarring, 23.5% for both hyperpigmentation and hypopigmentation (p<0.001 for scarring, p=0.017 for hyperpigmentation and p=0.022 for hypopigmentation).

Correlation of baseline xanthelasma characteristics with laser excision outcomes

Table 3 presents the results from logistic regression analyses investigating the relationship between various lesion characteristics and the success of laser excision. The analysis revealed a significant association between larger lesion heights and increased recurrence rates (OR 2.964, 95% CI 1.773 to 4.956; p<0.001). Conversely, the analysis indicated that recurrence risk was not significantly influenced by patients’ baseline age, gender, lesion location, grades or the number of treatment sessions (all p>0.05).

Table 3

Multivariable logistic analyses for the associations of baseline parameters with the risk of recurrence

Case illustrations

The following case examples illustrate the outcomes of CO₂ laser treatment for xanthelasma palpebrarum, with accompanying preoperative and postoperative images (figure 2).

Figure 2

Case examples of grade I–IV xanthelasma palpebrarum treatment. Case example of grade I lesion: (A) Pretreatment photograph of a 62-year-old female patient with grade I xanthelasma on both upper eyelids; (B) Post-treatment photograph taken 12 months after a single CO₂ laser excision session, showing complete lesion clearance with no recurrence or complications. Case example of grade II lesion: (C) Initial presentation of a 50-year-old female patient with grade II xanthelasma on both upper eyelids and the canthus; (D) Outcome 12 months post-treatment, following three CO₂ laser excision sessions, demonstrating total clearance without any signs of recurrence or complications. Case example of grade III lesion: (E) A 55-year-old female patient with grade III xanthelasma affecting the medial aspects of both upper and lower eyelids before treatment; (F) The appearance of the eyelids 12 months after completing three CO₂ laser excision sessions, with complete clearance and no evidence of recurrence or complications. Case example of grade IV lesion: (G) Preoperative image of a 41-year-old female patient with extensive grade IV xanthelasma on the medial and lateral aspects of both upper and lower eyelids; (H) Postoperative image captured 12 months after four CO₂ laser excision sessions, showing 95% clearance of the lesions with no recurrence or complications noted. CO2, carbon dioxide.

Case 1: A 62-year-old female patient presented with bilateral grade I xanthelasma on the upper eyelids, measuring 6×3 mm on the right and 5×3 mm on the left, with a maximum lesion height of 0.5 mm (figure 2A). She underwent a single session of CO₂ laser excision. At the 1-year follow-up, the patient exhibited complete lesion resolution, with no evidence of recurrence or postoperative complications (figure 2B).

Case 2: A 50-year-old female patient with grade II xanthelasma affecting both upper eyelids and the canthus was treated. The lesions measured 20×10 mm on the right and 27×11 mm on the left, with a maximum height of 2.5 mm (figure 2C). After three CO₂ laser sessions, the patient achieved full clearance, with no recurrence or complications noted at the 12-month post-treatment evaluation (figure 2D).

Case 3: A 55-year-old female patient was diagnosed with grade III xanthelasma, located on the medial aspects of both upper and lower eyelids. The lesion dimensions were 10×12 mm on the right upper eyelid, 10×2 mm on the right lower eyelid, 10×8 mm on the left upper eyelid and 17×5 mm on the left lower eyelid, with a lesion height of 1 mm (figure 2E). The patient received three sessions of CO₂ laser excision. Complete clearance was confirmed at the 12-month follow-up, with no signs of recurrence or complications (figure 2F).

Case 4: A 41-year-old female patient with grade IV xanthelasma exhibited lesions on both the medial and lateral aspects of the upper and lower eyelids. The lesion sizes were 25×10 mm on the right upper eyelid, 30×12 mm on the right lower eyelid, 27×14 mm on the left upper eyelid and 15×12 mm on the left lower eyelid, with a maximum height of 2 mm (figure 2G). Following four sessions of CO₂ laser treatment, the patient showed a 95% reduction in lesion size at the 1-year mark, with no recurrence or complications observed (figure 2H).

Discussion

Our investigation provides a comprehensive evaluation of CO₂ laser excision for large xanthelasma palpebrarum lesions. The technique has proven effective, with a high rate of lesion clearance and a manageable number of treatment sessions required for satisfactory outcomes. Postoperative complications were infrequent and generally mild, including hypertrophic scarring, hyperpigmentation and hypopigmentation.

Laser excision outperforms laser ablation for smaller lesions (grades I–III), with a clearance rate exceeding 99% and a recurrence rate of 6.8% across all grades, significantly lower than the recurrence rates associated with laser ablation (50% reported by Li et al13 and 20% reported by Pathania and Chatterjee).14 For large lesions, CO₂ laser excision rivals traditional surgery (8% recurrence rate reported by Elabjer et al7 and 25% complication rate reported by Choi et al19), offering comparable recurrence rates but with fewer complications.

We have demonstrated the advantages of laser excision over traditional methods. Laser excision minimises the risk of scar formation, hyperpigmentation and hypopigmentation, with incidences of 4.4%, 8.1% and 8.5%, respectively. The technique also avoids invasive reconstructive surgery, leading to a faster recovery, minimal bleeding and superior cosmetic outcomes. The procedure time is notably reduced, minimizing damage to surrounding tissues and enhancing patient comfort.

Despite the advantages of CO₂ laser excision for treating xanthelasma palpebrarum, certain limitations should be considered. First, larger xanthelasma lesions may require multiple treatment sessions due to the limited treatment area per session, potentially increasing the time and financial burden for patients. The multiple-session recommendation is reasonable, as larger wound sizes prolong the healing process, increasing inflammation and fibrosis and thereby elevating the risk of scarring.20 It is also important to note that wound healing occurred by secondary intention, with epithelialisation emerging in 1–2 weeks. Thus, close follow-up and proper wound care are especially important for patients with larger xanthelasma lesions to minimise scarring. Second, our study demonstrated that lesions with a height greater than 2 mm had a significantly higher probability of scarring and pigmentary changes compared with thinner lesions. Consistent to our study, Lee et al15 recommend considering surgical options for eyelid xanthelasma that surpass a height of 5 mm.

While our investigation benefits from considerable sample size and includes a wide range of extensive lesions, it is not without its constraints. The study’s single-centre, retrospective nature and the potentially short follow-up period may limit the generalisability of the findings. Future multicentre randomised controlled trials are necessary to corroborate our results and solidify the evidence base for the use of CO₂ laser excision in treating extensive xanthelasma palpebrarum.

In summary, CO₂ laser excision emerges as a precise, minimally invasive and effective treatment for large xanthelasma palpebrarum lesions. It offers significant advantages over traditional surgical and laser ablation methods, contributing to the evolving clinical evidence in favour of this modality as a preferred treatment option.

Supplemental material

Data availability statement

Data are available on reasonable request. The data including deidentified participant photographs are available on reasonable request (wangdq5@mail2.sysu.edu.cn).

Ethics statements

Patient consent for publication

Ethics approval

This study involves human participants and ethical approval was granted by the ethics committee of the Zhongshan Ophthalmic Center, Sun Yat-sen University, China (2023KYPJ286). Participants gave informed consent to participate in the study before taking part.

References

Footnotes

  • DW and ZM contributed equally.

  • Contributors XLin, YJ, and XLi contributed equally as co-corresponding authors. Conceptualisation: XLin and ZM; Methodology: DW and ZL; Investigation: WG and YQ; Writing–original draft: DW; Writing–review and editing: XLin and ZM; Resources: YQ, XLi and YJ; Supervision: XLin and ZM. XLin accepts full responsibility for the overall content as guarantors, ensuring the integrity of the research, access to the data, and control over the decision to publish.

  • Funding This work was supported by State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, Guangdong Provincial Key Laboratory of Ophthalmology and Visual Science, Guangdong Provincial Clinical Research Center for Ocular Diseases, Guangzhou, 510060, China.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.