Article Text

Download PDFPDF

Surgical options for lymphedema after gynecological cancer treatment: current trends and advances
Free
  1. Bernardo Pinheiro de Senna Nogueira Batista1 and
  2. Edward I-Fei Chang2
  1. 1 ACCamargo Cancer Center, Sao Paulo, São Paulo, Brazil
  2. 2 MD Anderson Department of Plastic Surgery, Houston, Texas, USA
    1. Correspondence to Dr Bernardo Pinheiro de Senna Nogueira Batista, ACCamargo Cancer Center, Sao Paulo, São Paulo, Brazil; bernardo.batista{at}accamargo.org.br

    Abstract

    Lower leg lymphedema is an important complication after gynecological treatment that can severely affect the quality of life of long-term survivors of these malignancies. As a chronic and progressive disease, affected patients will require life-long therapy centered on compression. Although conventional compressive treatments can be effective, they are extremely burdensome and time-consuming for most patients and adherence is challenging. With advances in the field of reconstructive microsurgery, new procedures have been developed in the past decades to help these patients in their continuous care and have been offered at many oncological centers around the world as a first line of treatment. We performed a PubMed search using the Mesh terms ‘Lymphedema/surgery’ and ‘Lower extremity’ yielding a total of 508 articles. Of these, 35 articles were included for analysis. Articles that failed to provide a comprehensive analysis of outcomes following surgical treatment, studies examining treatment for upper limb lymphedema, primary lymphedema, or lower extremity lymphedema resulting from non-gynecologic etiologies, and studies that failed to have a minimum of 6 months follow-up were excluded. A comprehensive review of these 35 articles including over 1200 patients demonstrated large variability on the outcomes reported; however, an overall benefit from these procedures was found. Surgical options including lymphovenous anastomosis, vascularized lymph node transfers, and excisional procedures can be performed in patients with lower leg lymphedema, depending on staging and findings in indocyanine green lymphography. Surgical treatment of lymphedema is an effective option that can improve symptoms and quality of life of patients suffering from lymphedema following gynecologic cancers.

    • Lymphatic System
    • Lymphatic Vessels
    • Postoperative complications
    • Gynecologic Surgical Procedures

    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.

    INTRODUCTION

    Lymphedema can be an important sequela after the treatment of gynecological cancers. The prevalence of lower leg lymphedema in this population has been estimated to be as high as 47.7% in the 10 years following the oncological extirpation including pelvic or inguinal node dissections, with body mass index higher than 25, and adjuvant radiotherapy constituting the major risk factors associated with its occurrence.1 For long-term cancer survivors, lower leg lymphedema is one of most debilitating and stigmatizing iatrogenic conditions following treatment.

    Lower leg lymphedema is a chronic and progressive disease, for which no cure is available. Clinically, the impairment of lymphatic fluid drainage will initially manifest as a swelling of the leg(s). If this is not addressed properly, the accumulation of the protein-rich and inflammatory lymph in the lymphatic collectors and subcutaneous tissues of the limb will lead to progressive damage of the local lymphatic system, with worsening of the clinical manifestations and increased resistance to clinical treatment. As the lymphatic fluid remains stagnant in the subcutaneous tissue, a progressive cycle of inflammation and fibrosis ensues that leads to increasing fat deposition and further compromise of the limb.

    The International Society of Lymphology (ISL) classifies lymphedema in four stages.2 In ISL stage 0, swelling is not clinically evident, but patients may already present with subjective symptoms, like limb heaviness, and show pathological patterns on lymphatic imaging studies. The inclusion of a pre-clinical stage of the disease highlights the importance of early diagnosis and treatment initiation to prevent disease progression. In ISL stage I, failure in fluid hemostasis by the lymphatic system becomes clinically evident with the accumulation of protein-rich fluid in the interstitial compartment that manifests as pitting edema with variable enlargement of the leg volume. In this initial stage, the swelling will subside with limb elevation, and patients will typically note resolution of the swelling on waking as the leg is elevated overnight, facilitating drainage from the legs due to the pressure gradient in the orthostatic position. In ISL stage II, pitting will still be present in the early phase, but the swelling no longer subsides with elevation and will be more resistant to compression. In late stage II, pitting edema may no longer be present, and the limb may further increase in size due to adipose hypertrophy and fibrosis of the subcutaneous tissue. Finally, in ISL stage III, also referred to as elephantiasis, the accumulation of fibroadipose tissue in the subcutaneous layer of the skin will lead to deformation of the leg.

    The impact of lower leg lymphedema in a patient’s quality of life cannot be understated. As the disease progresses, lymphedema will result in increased heaviness of the leg, progressive functional disability, and deterioration in the patient’s quality of life in terms of distress, depression, and autonomy reduction.3 4 Given the impaired lymphatic function in the leg, these patients are also at a higher risk of infection episodes that can require hospitalization and further accelerate deterioration of lymphatic collectors and disease progression.

    Traditionally, the gold standard of care for the management of lower leg lymphedema has been performed by specialized physiotherapists through complex decongestive therapy.5 This therapy is divided into two phases. In the initial active phase, patients are submitted to local skin care and intensive compressive treatment through multilayer bandaging. These sessions are typically performed three to five times a week over a course of several weeks, with the goal of removing the accumulated lymph and restoring the normal anatomy of the leg. In the second or passive phase, patients are transitioned to a compression garment to maintain the limb and prevent re-accumulation of fluid in the leg. These garments need to be used continuously and, for most patients, are insufficient to prevent progression. Consequently, these patients unfortunately will remain dependent on multiple compressive strategies to continuously control the swelling of their limbs and prevent disease progression. Complex decongestive therapy is an expensive and life-long treatment, and compliance is difficult for the majority of lower leg lymphedema patients, especially in hot, humid climates and resource deprived locations, as most of these treatments and garments are often not covered by insurance or healthcare systems. Additionally, in the later stages of the disease, as adipose and fibrotic tissue accumulates in the subcutaneous layer, compression alone will not be effective in restoring the normal anatomy of the affected leg.

    In the past decades, new surgical procedures have been developed to offer patients suffering from lower leg lymphedema additional options to treat their lymphedema. With the improvement in magnification and microsurgical instruments, associated with better lymphatic imaging, reconstructive surgery has expanded its applications into the field of lymphedema care. While still not a cure, these procedures have been performed worldwide, providing improvements in symptoms and quality of life of these patients. The purpose of this article is to perform a review of the current literature on their applications in patients with lower leg lymphedema secondary to gynecological cancer treatment.

    METHODS

    A PubMed search using the Mesh terms ‘Lymphedema/surgery’ and ‘Lower extremity’ was performed in September 2023. No other variable was used to limit search results. A total of 508 articles was found. Titles and abstracts were initially screened by a single reviewer to identify potential articles of interest for this review. Letters, case reports, reviews, articles with no abstract available, and those that included only patients with non-oncological lymphedema were excluded. Additionally, articles with outcomes reported with less than 6 months of follow-up after lymphatic surgical intervention were also excluded. A total of 87 articles were found to be eligible and selected for further analysis.

    The full text manuscripts of these 87 articles were searched using our institutional journal subscription database and were read by a single reviewer to determine final eligibility. Articles whose full text version were unavailable or written in languages other than English were also excluded. Additionally, articles that did not specify the number of lower leg lymphedema cases secondary to gynecological cancer treatment or had less than 25% of their subjects with this primary cause were also excluded (Figure 1).

    Figure 1

    PRISMA flow diagram. GC, secondary to gynecological cancer; PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

    A total of 35 articles met the inclusion and exclusion criteria described above and were included in the review. Each article was reviewed for the year of publication, the study design, which intervention(s) was being studied, the total number of patients and, of these, how many had lymphedema secondary to gynecological cancer treatment. During this review, the primary research outcome for each article and the lymphedema improvement outcomes reported were recorded. To determine the effect of treatment after surgery, the lymphedema improvement reported in each article was extracted.

    Given the tremendous variability in length of follow-up, reporting of outcomes, and methodology of measuring outcomes, data were extracted to provide the most consistent reporting of overall outcomes. Overall, the literature review provided a comprehensive overview of the different surgical treatment options for patients with lower leg lymphedema secondary to gynecological cancer treatment and their potential benefit for this population.

    RESULTS

    A total of 35 articles were included which comprised of 1 clinical trial, 23 case series and 11 cohort studies (Table 1). Most studies described the cumulative experience from different groups around the world or focused on technical improvements of the procedures. The number of articles focusing on lymphedema treatment has increased over the past decade, showing that these procedures, initially described in the 1990s, have been gaining popularity, and their availability for patients suffering with this severe complication from cancer treatment is rapidly expanding.

    Table 1

    Articles included for review

    The articles included analyzed the impact of three different surgical options that are commonly offered to patients with lymphedema. Lymphovenous anastomosis (21 articles),6–25 vascularized lymph node transfers (seven articles),26–32 and excisional procedures (seven articles)33–38 are the main procedures performed worldwide. The decision for which operation to perform varies according to lymphedema staging, clinical/imaging workup, and surgeon and patient preference, and is beyond the scope of this study. These procedures and their current indications will be presented and discussed in more detail in the next section.

    A single, small, clinical trial compared the association of lymphovenous anastomosis to compression, the gold standard treatment for lower leg lymphedema, and demonstrated a significant benefit in leg size decrease with improvement in patient symptoms and quality of life.6 Although the reported outcomes varied significantly among studies, they consistently focused on determining potential improvement in the subject’s objective and/or subjective symptoms. The reduction of the affected leg’s size was the most frequently reported outcome. The reduction in infectious events and cellulitis was reported in four articles and demonstrated a dramatic reduction in the number of cellulitis episodes between the years before and after surgery. Discontinuation of compressive therapy has also been reported in a few articles, and a series of different patient reported outcome measures specific to the lymphedema population have also been used to determine improvement.

    DISCUSSION

    In the modern era of lymphatic microsurgery, new procedures have been developed to assist lower leg lymphedema patients to improve the management of their condition and provide them an overall better quality of life. To that purpose, the introduction of indocyanine green (ICG) lymphography has allowed better visualization of the superficial lymphatics of the limb, with improved understanding of the underlying damage to the lymphatic system. ICG is a non-toxic, non-radioactive fluorescent dye that has been used in other fields of medicine for diagnostic angiography for decades. When injected subcutaneously in the interdigital spaces of the foot, it binds tightly to albumin and is captured and transported proximally by functioning lymphatic collectors. When excited by a laser with a specific wavelength, the dye absorbs and emits fluorescence that can be detected with an infrared camera. With ICG lymphography, superficial lymphatic collectors up to 1 cm in depth from the skin can be visualized and their function analyzed. Yamamoto et al classified the ICG lymphography findings in lower leg lymphedema patients into four stages that correlate to disease severity. In milder cases, the uptake and proximal migration of the ICG will show a linear pattern. This stage is often the most optimal for lymphovenous anastomosis as it identifies the location of a suitable lymphatic vessel that can be targeted for surgical intervention. As the disease progresses and the collectors lose their ability to transport the lymph, the dye extravasates into the interstitial space, showing a dermal backflow pattern that can be further classified into splash, stardust or diffuse according to progressive severity39 (Figure 2).

    Figure 2

    Indocyanine green lymphography showing linear pattern in the dorsum of the foot and distal third of the leg. Proximally, we can observe dermal backflow of the dye.

    Histologically, Mihara et al40 demonstrated histologically that lymphatic collectors will suffer degeneration as the disease progresses. With the increase in intraluminal lymph pressure, normal lymphatic vessels initially will become ectatic and have a larger diameter with thinner endothelial walls. Continuous pressure and inflammation will promote contraction of these vessels and their wall components will progressively degenerate into fibrotic elements. As these vessels degenerate to the contraction and sclerotic types, they will lose their ability to move lymph upward with stenosis of their lumen with eventual complete obstruction. These histological findings correlate with the previously described ICG lymphography stages.

    The MD Anderson ICG lymphography staging system41 classifies lymphographic findings into five stages in which progressively fewer patent lymphatic vessels (linear pattern) and impaired contractility are observed, with increasing severity of dermal backflow patterns along the limb. Clinically, this staging system has been used to classify potential candidates for lymphatic surgery to the most appropriate indication.

    Every patient with lower leg lymphedema is a potential candidate for surgery, but the indication will depend largely on clinical examination and ICG lymphography findings. A prerequisite for being considered for surgery, at least in our practice, is good adherence to compression and decongestive therapy. This is initially important to assess how much of the limb size is due to accumulated lymph and how much of it is due to a solid component, secondary to fibroadipose deposition. Other than correctly determining the ISL stage, this requisite also helps us select patients who can maintain an adequate compression protocol as this will be an important component of the post-operative care.

    Lymphovenous Anastomosis

    Lymphovenous anastomosis was initially described by Koshima in the 90s.42 The concept is based on the premise that creation of a functional anastomosis between subcutaneous lymphatic collectors and a recipient vein distal to the site of obstruction will divert the high-pressure lymph flow from the obstructed lymphatic system into a lower pressure venous system and alleviate the fluid accumulation in the affected limb. This procedure has the advantage of having very low morbidity, given the small incisions and superficial position of the structures used to perform the bypasses. There is no consensus on the post-operative care following lymphovenous anastomosis and is largely surgeon dependent. While some will hospitalize patients for 5 days following the operation, others prefer to perform it as an outpatient procedure under local anesthesia.

    The anastomoses are performed in submillimeter vessels, requiring more precise instruments and sutures than those used in conventional microsurgery. Lymphedema surgery has been termed supermicrosurgery to describe vessels <1 mm in size that are anastomosed typically using 11–0 or 12–0 nylon sutures. They are technically very challenging and require additional training from microsurgeons to perform functional anastomosis.

    Several configurations have been proposed for these anastomoses. Lymphatic collectors can be bypassed into subcutaneous veins through end-to-end (Figure 3) or end-to-side (Figure 4) techniques, with the literature showing a potential benefit from the latter15 23; however, more recent data seem to favor the former in terms of superior long-term patency of the anastomoses. An alternate technique that has emerged recently is the concept of side-to-end where the end of a vein is sutured to the side of a lymphatic vessel that may potentially favor drainage from the afferent and efferent sides of the collector, improving the efficacy of the bypass. Technically, this configuration is even more challenging and requires larger size vessels, which are not always found during dissection.

    Figure 3

    End to end lymphovenous anastomosis with nylon 11–0 (magnification x24). The patency of the anastomosis is confirmed by the flow of methylene blue from the lymphatic collector into the vein.

    Figure 4

    End to side lymphovenous anastomosis (magnification x24). The smaller lymphatic vessel has been sutured to the side of a larger subcutaneous vein using nylon 11–0.

    There is also no consensus on the ideal number of lymphovenous anastomoses to be performed to achieve the best response. Performing multiple anastomoses in the same procedure usually requires a multi-team approach, with more than one microscope and several supermicrosurgeons working simultaneously. In most settings, this is not feasible, and the number of anastomoses performed in a single procedure will most frequently be limited by the length of the procedure. In our practice, we will usually perform as many anastomoses as possible in a 3–4 hour setting. Additional lymphovenous anastomoses can be performed in a separate procedure if proven necessary. The favored location to perform these anastomoses is determined based on ICG lymphography at the sites where the linear pattern stops and dermal backflow of the ICG is observed. At these sites, we will normally find larger, ectatic lymphatic collectors that still have preserved contractile function to push lymph forward into the veins. Use of larger lymphatics has been associated with a higher success rate of performing functional anastomosis and better improvement outcomes.43

    Recently, advances in imaging such as magnetic resonance lymphoangiography and ultra-high frequency ultrasound allow for detection of lymphatics usually not visualized by ICG lymphography due to limitations in the depth of penetration. Identification of deeper lymphatics that may still be functional in advanced lower leg lymphedema allows the supermicrosurgeon to perform a lymphovenous anastomosis even in patients with more advanced lymphedema. Early outcomes have demonstrated promising results using this modality.19 However, in our practice, the lymphovenous anastomosis option is still reserved to the early initial MD Anderson stages 1 or 2, when we can still identify the presence of working lymph collectors in the affected limb by the presence of linear pattern in ICG lymphography.

    Given the low morbidity associated with the lymphovenous anastomosis, patients have minimal pain following the operation and a short recovery time. While there is still no consensus on when to resume compression and decongestive therapy, some surgeons will resume compression immediately following the operation. Regardless of when compression is initiated, patients can resume their daily routine sooner than the most invasive procedures available (Figure 5).

    Figure 5

    (A) Bilateral lymphovenous anastomoses (LVAs), immediately post-operatively. (B) Bilateral LVAs, 3 weeks post-operatively.

    Vascularized Lymph Node Transfer

    These procedures were initially described by Becker44 and consist of transferring functional lymph nodes from another part of the patient’s body to the lymphedematous leg. Technically, a flap consisting of adipose tissue from a region with a high concentration of lymph nodes is dissected and revascularized in the recipient site using conventional microsurgical techniques, with an arterial and venous anastomosis.

    The mechanism through which these transfers improve the lymphatic drainage of the limb has not yet been clarified. One hypothesis is that these functional, vascularized lymph nodes will work as physiological lymphovenous bypasses and drain part of the lymph accumulated in the leg. Others have suggested that these lymph nodes will stimulate formation of lymphatic vessels or lymphangiogenesis via high expression of specific endothelial growth factors in the affected limb.45 While the precise underlying mechanism remains to be elucidated, some lymphedema surgeons purport that vascularized lymph node transfers are more efficient than lymphovenous anastomosis in reducing lymph accumulation in the limb.26

    However, these are longer procedures with higher associated morbidity. Similar to conventional microsurgical free flaps, they require longer hospitalization and cannot be compressed in the initial weeks to prevent thrombosis of the vascular anastomosis. These patients will require a longer recovery time and will not be able to resume compression before at least 3 to 4 weeks. There is also a concern with the risk of secondary lymphedema at the donor site of these lymph node flaps, especially the inguinal and lateral thoracic donor sites. This is a rare but devastating complication that has been reported for all the donor sites except the submental lymph nodes and the intra-abdominal donor site.46 Technical advancements of the procedure have aimed to reduce this risk so we will always perform a double-staining technique or reverse mapping when harvesting these flaps. ICG is injected into the region of the donor site to detect lymph nodes to be incorporated into the flap, while methylene blue is injected distally in the foot or the hand to identify the most relevant sentinel nodes that must be avoided. Several donor sites have been proposed and include the inguinal, lateral thoracic, supraclavicular, and submental areas.

    Whenever possible, we will favor harvesting of lymph nodes of the omentum along the greater curvature of the stomach, based on the right gastroepiploic vessels. This can be done through a laparoscopic approach, that will require an additional team to perform the abdominal part of the operation.27 An additional benefit of this donor site is that the flap can usually be divided into two or more flaps and transferred to different sites of the affected leg (Figure 6).

    Figure 6

    Omental vascularized lymph node transfers (VLNTs), based on the right gastroepiploic vessels (clamps) harvested laparoscopically. The flap has been divided in two in a double VLNT transfer.

    In our practice, we perform the vascularized lymph node transfers in patients with intermediate MD Anderson stages 3 or 4 or when an initial lymphovenous anastomosis procedure failed to achieve the expected results. Vascularized lymph node transfers are also an alternative to improve lymphatic function in the legs of patients with a more severe condition, after excisional procedures have removed the excessive fibroadipose tissue.

    Excisional Procedures

    For more advanced cases of lower leg lymphedema, the chronic deposition of adipose and fibrotic tissue in the subcutaneous layer and terminal damage of the local lymphatic system will require more aggressive approaches to achieve satisfactory results. While intensive complex decongestive therapy and physiological procedures can still help reduce the accumulation of lymph, they will not improve the chronic deterioration and fibrosis of the leg already established. Therefore, these patients will require some form of debulking procedure to return their legs to a more anatomical shape.

    Suction-assisted lipectomy consists of circumferential liposuction of the affected leg, aimed at removing the excessive fibroadipose tissue deposited in these legs. This has been initially proposed by Brorson and Svensson47 and has been the most commonly performed procedure for later stages of lower leg lymphedema. These procedures carry a very high morbidity and improvement is highly dependent on adequate rehabilitation and compressive therapy for months or often permanently after the procedure to prevent fluid accumulation and intense fibrosis in the operated areas.

    Other more aggressive forms of excisional procedures are reserved for only the most extreme cases of lymphedema, so-called elephantiasis, and entail the entire resection of the skin and subcutaneous tissue to the level of the muscle fascia followed by skin grafting. This is known as the Charles’ procedure and is reserved for very advanced stages of lower leg lymphedema, given the extremely high morbidity and poor esthetic results. This procedure can be performed in combination with vascularized lymph node transfers with the intention of improving lymphatic fluid drainage.37

    Outcome Assessment

    Although the reduction of the affected leg’s size was the most frequently reported outcome in the articles included in this review, there is an important variability among the measuring strategies used. Circumferential measurements to determine limb girth pre- and post-operatively at one or multiple pre-determined sites is the simplest way of determining improvement. These outcomes can be expressed as an absolute or relative difference (limb circumference reduction and circumference reduction rate, respectively). Unfortunately, while many studies purport on the reliability of this technique, there is tremendous variability based on who is obtaining the measurements and the precise locations that are measured. A more complex form of analysis is to estimate the volume of the limb, using the truncated cone formula, where circumferential measurements are obtained at the perimeter of the upper and lower parts of the limb. Limb volume can also be assessed using a specific infrared device (perometer) or by estimation in MRI studies. The variation between pre- and post-operative limb volumes has also been expressed in absolute or relative differences (limb volume reduction and volume reduction rate).

    Unfortunately, these direct comparisons between pre- and post-operative measurements will fail to compensate for other important confounding factors, such as weight variation over time, that will certainly impact the size of both the affected and the unaffected limb in unilateral cases. Therefore, some authors will perform the same measurements in the contralateral healthy limb and determine the absolute or relative excess volume of the affected limb (excess volume and percentage of excess volume, respectively). The lower extremity lymphedema index has been proposed as an alternative measure to compensate for body mass index variations over time. It is determined by the sum of the square of five girth measurements along the leg, divided by the body mass index at the time of measurement.48

    These different strategies to determine the impact of treatment on the affected leg size can be misleading, and the variability of reported outcomes poses an additional challenge for interpretation of their findings. Furthermore, we have previously shown that these mathematical estimations based on measurements that are, by definition, subject to some degree of error, can increase the observational error of the estimated outcomes exponentially.49 Thus, while circumferential measurements are the simplest means of obtaining objective data regarding changes in limb volume and size, they are notoriously inaccurate in the authors’ opinion.

    The reduction of infectious episodes, another significant complication lymphedema patients encounter, has also been documented as a benefit reflecting the success of the operation. This outcome alone has a major impact on the lymphedema patient’s quality of life, reducing the need for hospitalization and costs to the patient and healthcare system, lifelong antibiotic suppression, and disease progression.

    Discontinuation of compressive therapy would ideally be considered the optimal outcome, but lymphedema is still an incurable disease, and these procedures should not be offered as a definitive cure. For most patients, compression will remain necessary to achieve the most optimal outcomes following lymphedema surgery. The patient’s dependence on compression is therefore a variable that can act as a confounding factor but could also be used as an important outcome for these procedures. Reducing the overall need for compression is one of the most valued results from surgery we observe in our practice.

    More recently, the use of patient-reported outcome measures specific to the lymphedema population has been incorporated into the treatment and management of lymphedema patients, and should provide broader and more meaningful tools for the evaluation of the effect of these procedures.50 These validated metrics have been implemented in several different surgical fields such as the BREAST-Q for patients undergoing reconstruction following breast cancer, and are designed to provide an objective assessment of the patient’s subjective experience and quality of life. The most widely used measures for lymphedema are the lymphedema quality of life (LYMQOL), lymphedema life impact scale (LLIS), lymphedema quality of life inventory (LyQLI), and the LYMPH-Q.

    CONCLUSION

    In the last decades, advances in the field of reconstructive microsurgery have renewed interest in the treatment of secondary lymphedema. Although it remains a chronic and progressive disease, different procedures can be offered to these patients, according to the severity of their presentation. Physiological (lymphovenous anastomosis and vascularized lymph node transfers) and excisional procedures have been shown to improve quality of life, reduce the size of affected legs and episodes of infection and, therefore, decelerate disease progression. For patients suffering from lymphedema, lymphedema surgery can provide important, life-changing benefits that should be considered with the improved survival and prognosis of patients who have undergone treatment for gynecologic malignancies. Despite the already established evidence on the benefits of these procedures, more research is needed to advance the field and define proper patient selection, post-operative management, and establish universal outcomes assessment.

    Ethics statements

    Patient consent for publication

    Ethics approval

    Not applicable.

    References

    Footnotes

    • Contributors Both authors contributed to the final manuscript.

    • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

    • Competing interests None declared.

    • Provenance and peer review Commissioned; externally peer reviewed.