Legg Calve Perthes disease: a retrospective study of pelvic radiographic parameters

doi:10.21203/rs.3.rs-3865841/v1

Download PDF

Research Article

Legg Calve Perthes disease: a retrospective study of pelvic radiographic parameters

https://doi.org/10.21203/rs.3.rs-3865841/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 20 Sep, 2024

Read the published version in BMC Musculoskeletal Disorders →

You are reading this latest preprint version

Purpose

Legg Calve Perthes disease (LCPD) is a paediatric hip disorder caused by ischemia of the femoral epiphysis, sometimes causing femoral head deformity. This study aims to determine if previously validated pelvic obliquity in skeletally immature patients through radiographic parameters for assessing acetabular retroversion in developmental dysplasia of the hip are applicable to patients with LCPD and its prognostic value.

Methods

A retrospective observational study of patients with Legg Calve Perthes disease was carried out, in which 4 pelvic parameters were studied: Ilioischial Angle, Obturator Index, Sharp’s Angle and Acetabular Depth-Width Ratio (ADR), analysing the differences between the healthy and affected hips. and subsequently, it is assessed whether the differences in these parameters may have a prognostic value in the outcome of the disease.

Results

Statistically significant differences have been obtained in the ilioischial angle, obturator index and ADR, between the affected and healthy hip. But regarding its value in the prognosis of the disease, only the Acetabular Depth Width Ratio has shown to have a predictive value of the outcome of the disease.

Conclusions

The present study showed that the pelvic parameters studied differ between the healthy and the diseased hip, but only one of them, the ADR, seems to have significance on the evolution and prognosis of the disease. More studies with a larger sample size are necessary, but we believe that studies on pelvic obliquity may open a new line of research in Legg Calve Perthes disease.

In 1910, Arthur T. Legg, Jacques Calve, and Georg Clemenc Perthes independently documented a paediatric hip disorder arising from ischemia affecting the proximal femoral epiphysis [1], mainly in males between 4 and 8 years old. This leads to avascular necrosis in the affected region resulting in a clinical presentation characterized by pain, limited mobility, and claudication. Additionally, the initial necrosis of the femoral head involves a series of progressive changes, transitioning from initial ischemia with sclerosis to fragmentation, and finally, revascularization with bone neoformation, with eventual healing, as per the Waldenstrom classification depicted in Fig. 1 [2]. The ultimate outcome varies depending on the extent of deformity in the femoral head and the achieved articular congruence at the conclusion of the process.

The femoral head changes resulting from necrosis have been extensively studied, with various classifications such as Catterall, Salter and Thomson, and more recently, Herring's lateral pillar classification. Additionally, concerning the long-term outcomes related to the femoral head's sphericity and the achieved articular congruence at the process's conclusion, Stulbert's classification was developed [3] (see Table).

In this classification, grade I represents a completely normal hip joint. In Grade II, there is spherical congruence, albeit with minimal deformity of the femoral head, measuring less than 2 mm. Grade III indicates non-spherical congruence, with a loss of femoral head shape exceeding 2 mm. Grade IV involves a flattened femoral head with abnormalities in the femoral head and neck but congruence in the acetabulum. Lastly, Grade V signifies non-spherical incongruence, presenting a flattened femoral head with acetabular incongruence. In addition to the previously mentioned femoral head deformity, radiographic studies have identified acetabular retroversion and pelvic obliquity. Many radiographic studies have also noted asymmetry in both obturator foramina, resembling a "Chinese eye" as illustrated in Fig. 2.

This is significant because the acetabulum plays a crucial role during the remodelling phase of the femoral head. The unchanged shape of the acetabulum facilitates the restoration of the sphericity of the proximal femoral epiphysis. However, alterations in the acetabulum's shape and inadequate coverage of the deformed femoral head can potentially lead to early onset osteoarthritis of the hip.

As LCPD typically manifests before reaching skeletal maturity, often between 4 and 8 years of age [4], accurately evaluating the shape of the acetabulum through radiographs during these ages is challenging due to incomplete ossification. Existing literature on acetabular retroversion in LCPD primarily focuses on skeletally mature hips with healed LCPD [5, 6, 7]. It is essential to note that acetabular retroversion involves not just the acetabulum but the entire hemipelvis [8, 9].

While CT or MRI can assess acetabular retroversion in the early stages of LCPD [10], these scans are not routine due to CT's radiation and the potential need for sedation or anaesthesia with MRI. Consequently, new indirect radiographic parameters have been proposed to evaluate acetabular retroversion before acetabular ossification [11].

This study aims to determine if previously validated radiographic parameters for assessing acetabular retroversion in developmental dysplasia of the hip (DDH) [12] are applicable to patients with LCPD. The objective is to identify potential variations between healthy hips and those affected by LCPD. Additionally, the study aims to investigate whether these parameters can serve as reliable factors in predicting the prognosis of LCPD and if they correlate with the outcome according to the Stulberg’s classification.

A retrospective observational study was conducted on patients diagnosed with LCPD who received care at a Children's Orthopaedic Unit of the Santiago University Clinical Hospital from 2014 to 2021. The study involved collecting demographic variables and reviewing radiographic studies in both anteroposterior and axial projections. Only patients with unilateral involvement were included, and they were followed up until reaching skeletal maturation. The affected hip was analysed, with the healthy hip serving as the control group.

All radiographs were consistently obtained following a standardized protocol. The patient assumed a supine decubitus position, with a film focusing distance of 1.2 m and the beam centred between the pubic symphysis. The pelvis was positioned in neutral rotation, confirmed by verifying correct longitudinal rotation when the tip of the coccyx aligned with the pubic symphysis.

In the radiographic assessments, the extent of femoral head involvement was evaluated using Herring's classification. Additionally, the presence of signs indicating a femoral head at risk was examined, and assessments of acetabular retroversion and pelvic obliquity were conducted by examining specific pelvic parameters. These parameters, previously validated by other authors in the context of DDH sequelae, include the ilioischial angle.

The ilioischial angle is determined by drawing two lines. The first line connects the base of each acetabular teardrop, forming the "interteardrop line." The second line extends from the intersection of the ilioischial line and the iliopectineal line to the outermost point of the ipsilateral obturator foramen. The measured angle is the one formed between these two lines, as illustrated in Fig. 3.

The obturator index is determined by delineating two lines. The first line extends across the maximum width of the obturator foramen (c), while the second is half the length of the interteardrop line (d). The index is computed by dividing the length of the first line by half the length of the interteardrop line, as illustrated in Fig. 4.

Sharp's angle is formed by a line drawn between the two inferior points of both teardrops and another line connecting the inferior point of one teardrop to the lateral edge of the acetabulum, as depicted in Fig. 5.

The acetabular depth-to-width ratio (ADR) is determined by a line connecting the superior bony margin of the acetabulum to the inferior margin of the ipsilateral teardrop (width measurement). Acetabular depth is defined by a line extending from the width line to the deepest point of the acetabular cavity. Once these measurements are acquired, the acetabular depth is divided by the width and then multiplied by a thousand, following the formula: (depth/width) x 100, as shown in Fig. 6.

For consistency, these evaluations were conducted on all patients in Waldenström stage II, as acetabular retroversion is notably prevalent in this stage [11].

The radiographic measurements, anonymized for privacy, were initially taken in both hips by the same senior specialist in Orthopaedic Surgery. A revaluation was carried out a month later by the same reviewer. Following the receipt of Ethics Committee approval and the acquisition of informed consents, the radiographic studies of 65 patients were examined. However, 21 were excluded after applying inclusion criteria—20 due to imperfectly centred radiographs or interference with measurements caused by the genital protector, and 1 due to discrepancies in measurements by the reviewer. Consequently, the study was completed with 44 patients.

Statistical analysis was conducted using IBM SPSS Statistics, version 22, and Microsoft Excel. A descriptive analysis of pelvic parameters was executed, encompassing mean, standard deviation, standard error, and a 95% confidence interval for the corresponding mean of the dependent variable for each studied parameter across both hips.

Initially, a comparison was made between the results obtained for each pelvic obliquity parameter in affected hips and those unaffected by the pathology. Subsequently, the suitability of these parameters in predicting the disease prognosis was assessed by comparing them with the final hip condition according to the Stulberg’s classification, utilizing the ANOVA statistical test.

A significance level of 0.05 was employed, signifying that differences were considered significant when the p-value fell below 0.05. The F values for each pelvic parameter were examined to determine whether statistical significance was greater or less than 0.05. Consequently, the null hypothesis of equality of means was accepted if the significance was greater, or conversely, rejected if the significance was less than 0.05, thereby accepting the alternative hypothesis. This study has been approved by the CEI-SL (Comité de ética de la Investigación Santiago Lugo) ethics committee (Code 2023/326) and have therefore been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki. All persons gave their informed consent prior to their inclusion in the study. Since subjects are minor, informed consents have been obtained from parents or legal representative. Details that might disclose the identity of the subjects under study have been omitted.

The age range of the patients varied from two years and nine months to seven years and eight months, with a mean age of five years and three months (1933.52 days) and a standard deviation of one year, seven months, and seventeen days (592.26 days). Out of the patients, only two were female, constituting 9.52%.

Regarding the degree of involvement, as per the Herring classification for femoral head involvement, 10 corresponded to group A, 26 to type B, and 8 to type C.

Statistically significant differences in the ilioischial angle, obturator index, and ADR were observed when comparing pelvic parameter measurements between healthy hips and those affected by LCPD. However, no significant difference was found in Sharp's angle (refer to Table 1).

Table 1 Statistiscal analysis summary of pelvic parameters. ADR acetabular width-to-depth ratio, CI Confidence Interval, Sign. Significance after Shapiro-Wilk test.

ilioischial angle

obturator index

Sharp’s angle

ADR

LCPD hip

Mean 95,5° +/- 3,4°

CI 95% 94-96,6

Variance 11,6

Sign 0,30

Mean 0,43 +/- 0,57

CI 95% 0,42 – 0,46

Variance 0,003

Sign 0,42

Mean 44,7° +/- 3,9

CI 95% 43°– 46°

Variance 15,9

Sign 0,74

Mean 278,3 +/- 32,7

CI 95% 264-332

Variance 1169,7

Sign 0,89

Healthy hip

Mean 93,1° +/-3,2°

CI 95% 91,6-94,7

Variance 10,5

Sign 0,4

Mean 0,51+/- 0,34

CI 95% 0,49-0,52

Variance 0,001

Sign 0,5

Mean 43,7° +/-2,6

CI 95% 43,1°- 45,7°

Variance 6,7

Sign 0.05

Mean 306,7 +/- 32,7

CI 95% 292-322

Variance 1074,1

Sign 0,22

Tabl2 2 . ANOVA of pelvic parameters according to Stulberg’s classification

	Source of variation	Root mean square	F index	Statistical Significance
ilioischial angle	Intergroups Intragroups	2,57 14,47	0,179	0,93
obturator index	Intergroups Intragroups	0,004 0,003	1,16	0,36
Sharp’s Angle	Intergroups Intragroups	10,203 17,485	0,58	0,60
ADR	Intergroups Intragroups	2509,686 818,523	3,06	0,04

An analysis was conducted to determine if the values of these pelvic parameters could impact the prognosis of the disease according to the Stulberg classification. Among the 44 hips, 4 were classified as Grade I, 5 as Grade II, 13 as Grade III, 15 as Grade IV, and 7 as Grade V.

Examining the results presented in Table 2, it is evident that there are no statistically significant differences, given that the p-value is greater than 0.05, except for the grades and ADR. This implies that the null hypothesis is accepted for the first three parameters in the table—namely, the ilioischial angle, obturator index, and Sharp angle have independent values concerning the assigned Stulberg grade of the affected hips. On the other hand, the acetabular width-to-depth ratio (ADR) exhibits statistical significance less than 0.05. Consequently, the values of this variable fluctuate concurrently with the Stulberg classification, suggesting prognostic value in the progression of LCPD.

Clinical symptoms, disease onset age, and radiological signs play crucial roles in determining the final clinical and functional outcomes following hip joint treatment for Perthes disease. Numerous imaging studies have focused on deformities in hips affected by Legg-Calvé-Perthes disease (LCPD), primarily emphasizing alterations in the femoral head. The risk head concept proposed by Catterall has been widely utilized to estimate hip joint form and function during LCPD treatment. Different classifications, such as those by Catterall, Salter and Thomson, and Herring, have been developed based on femoral head involvement, with Herring's classification currently being highly regarded as a prognostic factor. However, acetabular retroversion, present in 31–49% of LCPD cases, is also observed in these imaging studies [18, 19].

Acetabular retroversion further limits range of motion and exacerbates femoroacetabular impingement in LCPD hips. The biomechanics of the hip reveal an intimate relationship between the shape of the femoral head and pelvic morphology [20]. Acetabular retroversion significantly influences femoral head coverage and the orientation of the lateral wall of the acetabulum. Moreover, the acetabulum plays a pivotal role in the remodelling of the proximal femoral epiphysis and, consequently, in the ultimate outcome of the disease.

For the evaluation of acetabular retroversion, as far as imaging tests are concerned, the ossification of the acetabulum plays an important role, but complete closure of the triradiate cartilage usually occurs in women and men at the age of 15 and 16 years, respectively [22].

Assessing acetabular retroversion in skeletally immature hips with Perthes disease poses challenges due to the incomplete ossification of the acetabulum, limiting standard radiographic evaluation. Consequently, CT and MRI studies have been employed, albeit with concerns about radiation exposure and the potential need for sedation or anaesthesia in children.

Despite limited reliable studies on acetabular retroversion in skeletally immature hips with Perthes disease, contradictory findings in the early stage (Waldenström stage I) shift in prevalence and image homogeneity in in stages II or III with fragmentation and new bone formation, as well as a greater homogeneity of the images, in these stages (Fig. later stages (Fig. 5). Various radiographic parameters, such as the crossover sign [23], retroversion index [24], and posterior wall sign [5], have been proposed to assess acetabular retroversion, considering the limitations of evaluating the ischial spine sign in immature hips.

Contrary to the affected hip, the contralateral side (unaffected hip) rarely exhibits changes in acetabular version over the same period.

In Leibold's study [24]. They appreciated that the greater the severity of the affectation of the femoral epiphysis Herring B or C was accompanied by greater acetabular retroversion. Likewise, in Larson's study [22], he speculates on the hypothesis that acetabular retroversion by causing a femoroacetabular impingement could cause microtrauma in the retinacular vessels, which would have etiological importance in LCPD, although this speculation has several limitations due to the inclusion criteria.

This study evaluates pelvic parameters in Perthes disease, focusing on their possible prognostic value rather than surgical treatments. The pelvis's orientation during radiograph acquisition directly impacts radiographic anatomy, leading to the exclusion of improperly centered or oriented radiographs. Measurements, performed by a single observer, were repeated after one month to minimize inter-observer differences.

While inclination may affect radiographic measurements, studies indicate its similar impact on both hips. Despite the study's limitations, statistically significant differences were observed in the ilioischial angle, obturator index, and acetabular depth-width ratio between healthy and PCL-affected hips in the initial stages of Waldestron stage II disease, offering insights beyond studies requiring skeletal maturation for pelvic obliquity assessment.

Few studies have explored the relationship between final outcomes and pelvic and acetabular obliquity, making Grzegorzewski's [25] evaluation of acetabulum shape changes during the disease process particularly interesting. Notably, acetabula with a normal shape at the disease's conclusion were classified as Stulberg's group 1 and 2, while those with deformities in the acetabulum's external portion were classified as Stulberg's type 4 and 5, with no explicit reference to pelvic obliquity.

In our investigation, patients with PCCL exhibited distinct values for the mean obturator index, ilioischial angle, and ADR between the affected and healthy hips, whereas Sharp's angle showed no such difference. This suggests the presence of pelvic obliquity in the affected hip compared to the healthy hip in LCPD, as assessed during measurements conducted in Waldenström phase II.

However, concerning the significance of these measurements and their impact on the disease prognosis, only the ADR has demonstrated a correlation with the final Stulberg’s classification.

a. Ethics approval and consent to participate

This study has been approved by the CEI-SL (Comité de ética de la Investigación Santiago Lugo) ethics committee (Code 2023/326) and have therefore been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki. All persons gave their informed consent prior to their inclusion in the study. Details that might disclose the identity of the subjects under study have been omitted.

b. Consent for publication

Not Applicable

c. Availability of data and materials

The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.

d. Competing interests

We declare that the authors have no competing interests as defined by BMC, or other interests that might be perceived to influence the results and/or discussion reported in this paper.

e. Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

f. Authors' contributions

MVG Acquisition and data, Drafting of the manuscript CTP Acquisition and data, Drafting of the manuscript LPL, Drafting of the manuscript, Acquisition and data YF Analysis and interpretation of data, Statistical analysis RFP Analysis and interpretation of data, Acquisition and data JCA Analysis and interpretation of data, Acquisition and data OG Conception and design, Analysis and interpretation of data, Statistical analysis, Critical revision of the manuscript for important intellectual content JP Conception and design, Analysis and interpretation of data, Statistical analysis, Critical revision of the manuscript for important intellectual content We certify that each author has participated in and has contributed sufficiently to the work to take public responsibility for the appropriateness of the experimental design and method, and the collection, analysis and interpretation of the data and that this final version has been reviewed and approved for submission and/ or publication. We also certify that the sequence of authorship below is identical to that on the submitted manuscript.

Leroux J, Abu Amara S, Lechevallier J. Legg-Calvé-Perthes disease. Orthop Traumatol Surg research: OTSR. 2018;104(1S):107–S112. https://doi.org/10.1016/j.otsr.2017.04.012.
Perry DC, Bruce CE, Pope D, Dangerfield P, Platt MJ, Hall AJ. Perthes' disease of the hip: socioeconomic inequalities and the urban environment. Arch Dis Child. 2012;97(12):1053–7. https://doi.org/10.1136/archdischild-2012-302143.
García Mata S, Ardanaz Aicua E, Hidalgo Ovejero A, Grande M, M. Legg-Calvé-Perthes disease and passive smoking. J Pediatr Orthop. 2000;20(3):326–30.
Wiig O, Terjesen T, Svenningsen S, Lie SA. The epidemiology and aetiology of Perthes' disease in Norway. A nationwide study of 425 patients. J Bone Joint Surg Br Vol. 2006;88(9):1217–23. https://doi.org/10.1302/0301-620X.88B9.17400.
Vosmaer A, Pereira RR, Koenderman JS, Rosendaal FR, Cannegieter SC. Coagulation abnormalities in Legg-Calvé-Perthes disease. J Bone Joint Surg Am Vol. 2010;92(1):121–8. https://doi.org/10.2106/JBJS.I.00157.
Hresko MT, McDougall PA, Gorlin JB, Vamvakas EC, Kasser JR, Neufeld EJ. Prospective reevaluation of the association between thrombotic diathesis and legg-perthes disease. J Bone Joint Surg Am Vol. 2002;84(9):1613–8. https://doi.org/10.2106/00004623-200209000-00014.
Sirvent N, Fisher F, el Hayek T, Appert A, Giudicelli H, Griffet J. Absence of congenital prethrombotic disorders in children with Legg-Perthes disease. J Pediatr Orthop B. 2000;9(1):24–7. https://doi.org/10.1097/01202412-200001000-00005.
Lee JH, Zhou L, Kwon KS, Lee D, Park BH, Kim JR. Role of leptin in Legg-Calvé-Perthes disease. J Orthop research: official publication Orthop Res Soc. 2013;31(10):1605–10. https://doi.org/10.1002/jor.22415.
Miyamoto Y, Matsuda T, Kitoh H, Haga N, Ohashi H, Nishimura G, Ikegawa S. A recurrent mutation in type II collagen gene causes Legg-Calvé-Perthes disease in a Japanese family. Hum Genet. 2007;121(5):625–9. https://doi.org/10.1007/s00439-007-0354-y.
The classic. The first stages of coxa plana by Henning Waldenström. 1938. (1984). Clinical orthopaedics and related research, (191), 4–7.
McAndrew MP, Weinstein SL. A long-term follow-up of Legg-Calvé-Perthes disease. J Bone Joint Surg Am Vol. 1984;66(6):860–9. https://doi.org/10.2106/00004623-198466060-00006.
Wang L, Bowen JR, Puniak MA, Guille JT, Glutting J. An evaluation of various methods of treatment for Legg-Calvé-Perthes disease. Clin Orthop Relat Res. 1995;314:225–33.
Stulberg SD, Cooperman DR, Wallensten R. The natural history of Legg-Calvé-Perthes disease. J Bone Joint Surg Am Vol. 1981;63(7):1095–108.
Froberg L, Christensen F, Pedersen NW, Overgaard S. Long-term follow-up of a patient cohort with Legg-Calvé-Perthes disease. J Pediatr Orthop B. 2011;20(5):273–7. https://doi.org/10.1097/BPB.0b013e3283474268.
Herring JA, Kim HT, Browne R. Legg-Calve-Perthes disease. Part I: Classification of radiographs with use of the modified lateral pillar and Stulberg classifications. J Bone Joint Surg Am Vol. 2004;86(10):2103–20.
Herring JA, Kim HT, Browne R. Legg-Calve-Perthes disease. Part II: Prospective multicenter study of the effect of treatment on outcome. J Bone Joint Surg Am Vol. 2004;86(10):2121–34.
Herring JA, Neustadt JB, Williams JJ, Early JS, Browne RH. The lateral pillar classification of Legg-Calvé-Perthes disease. J Pediatr Orthop. 1992;12(2):143–50. https://doi.org/10.1097/01241398-199203000-00001.
Leibold CS, Whitlock P, Schmaranzer F, Ziebarth K, Tannast M, Steppacher SD. Development of acetabular retroversion in LCPD hips-an observational radiographic study from early stage to healing. Arch Orthop Trauma Surg. 2023;143(7):3945–56. https://doi.org/10.1007/s00402-022-04612-0.
Tannast M, Pfannebecker P, Schwab JM, Albers CE, Siebenrock KA, Büchler L. Pelvic morphology differs in rotation and obliquity between developmental dysplasia of the hip and retroversion. Clin Orthop Relat Res. 2012;470(12):3297–305. https://doi.org/10.1007/s11999-012-2473-6.
Hogervorst T, Bouma H, de Boer SF, de Vos J. Human hip impingement morphology: an evolutionary explanation. J Bone Joint Surg Br Vol. 2011;93(6):769–76. https://doi.org/10.1302/0301-620X.93B6.25149.
Sankar WN, Flynn JM. The development of acetabular retroversion in children with Legg-Calvé-Perthes disease. J Pediatr Orthop. 2008;28(4):440–3. https://doi.org/10.1097/BPO.0b013e318168d97e.
Larson AN, Stans AA, Sierra RJ. Ischial spine sign reveals acetabular retroversion in Legg-Calvé-Perthes disease. Clin Orthop Relat Res. 2011;469(7):2012–8. https://doi.org/10.1007/s11999-011-1793-2.
Tannast M, Pfannebecker P, Schwab JM, Albers CE, Siebenrock KA, Büchler L. Pelvic morphology differs in rotation and obliquity between developmental dysplasia of the hip and retroversion. Clin Orthop Relat Res. 2012;470(12):3297–305. https://doi.org/10.1007/s11999-012-2473-6.
Leibold CS, Whitlock P, Schmaranzer F, Ziebarth K, Tannast M, Steppacher SD. Development of acetabular retroversion in LCPD hips-an observational radiographic study from early stage to healing. Arch Orthop Trauma Surg. 2023;143(7):3945–56. https://doi.org/10.1007/s00402-022-04612-0.
Grzegorzewski A, Synder M, Kozłowski P, Szymczak W, Bowen RJ. The role of the acetabulum in Perthes disease. J Pediatr Orthop. 2006;26(3):316–21. https://doi.org/10.1097/01.bpo.0000221926.10148.bf.

No competing interests reported.

Download PDF

Journal Publication

published 20 Sep, 2024

Read the published version in BMC Musculoskeletal Disorders →

Editorial decision: Revision requested
29 Feb, 2024
Reviews received at journal
15 Feb, 2024
Reviewers agreed at journal
15 Feb, 2024
Reviewers invited by journal
15 Feb, 2024
Editor assigned by journal
15 Feb, 2024
Editor invited by journal
26 Jan, 2024
Submission checks completed at journal
26 Jan, 2024
First submitted to journal
15 Jan, 2024

You are reading this latest preprint version

Legg Calve Perthes disease: a retrospective study of pelvic radiographic parameters

Status:

Journal Publication

Version 1

Abstract

Purpose

Methods

Results

Conclusions

Figures

INTRODUCTION

MATERIAL AND METHODS

RESULTS

DISCUSSION

CONCLUSIONS

Declarations

References

Additional Declarations

Status:

Journal Publication

Version 1