|Ahead of print
Power Doppler sonography – A supplement to hysteroscopy in abnormal uterine bleeding: Redefining diagnostic strategies
Shuchita Batra1, Anuradha Khanna1, RC Shukla2
1 Obstetrics and Gynaecology, Institute of Medical Sciences, Banaras Hindu University, Varanasi, Uttar Pradesh, India
2 Radiodiagnosis, Institute of Medical Sciences, Banaras Hindu University, Varanasi, Uttar Pradesh, India
|Date of Submission||25-Jul-2019|
|Date of Decision||25-Jan-2020|
|Date of Acceptance||13-Apr-2020|
|Date of Web Publication||21-Mar-2021|
Obstetrics and Gynaecology, Institute of Medical Sciences, Banaras Hindu University, Varanasi, Uttar Pradesh
Source of Support: None, Conflict of Interest: None
Background: Abnormal Uterine Bleeding (AUB) is a very frequent cause of gynecological visits in women of all age groups. Ultrasound pelvis with or without endometrial sampling have been conventionally used to make diagnosis. Power Doppler is a comparatively recent modality which can be used to screen patients who will need endometrial biopsy/ curretage. We hereby conducted a study to compare the diagnostic accuracy of power Doppler sonography and hysteroscopy with histopathology associated with abnormal uterine bleeding. We also calculated the incidence of uterine pathology in AUB by power Doppler ultrasound and hysteroscopy and compared it with histopathology.
Methods: This prospective cohort study was conducted at the Department of Obstetrics and Gynaecology, Institute of Medical Sciences, Banaras Hindu University. After excluding 42 women, a total of 100 women fulfilling the inclusion criteria contributed to our study. Selected women underwent power Doppler ultrasound and hysteroscopy with guided biopsy. Results were compared with histopathology as per the gold standard. Evaluation of sensitivity, specificity, positive and negative predictive values were performed for each modality. All statistical analyses were performed using the SPSS 11.0 statistical package. P value ≤0.05 was considered statistically significant for all tests used.
Results: Sensitivity and specificity of power Doppler are 75% and 100% for carcinoma endometrium, 72.72% and 98.9% for endometrial hyperplasia, and 81.81% and 100% for endometrial polyp, respectively.
Conclusion: Power Doppler sonography can be used to screen outpatients who do not need an endometrial biopsy for abnormal uterine bleeding. This will avoid unnecessary hysteroscopy in definitive benign cases, and watchful hysteroscopy in suspected premalignant and malignant cases. Irregular branching vessels and color splashes were found to be the best parameters for diagnosing endometrial carcinoma. Power Doppler should be done along with transvaginal sonography in all cases of abnormal uterine bleeding.
Keywords: Abnormal uterine bleeding, endometrial carcinoma, international endometrial tumor analysis (IETA), power Doppler sonography, hysteroscopy
Key Message Widespread use of power Doppler sonography will help in reducing the unnecessary hysteroscopies and dilatation and curretages in patients with abnormal uterine bleeding. This will provide early diagnosis and reduce the cost of treatment.
|How to cite this URL:|
Batra S, Khanna A, Shukla R C. Power Doppler sonography – A supplement to hysteroscopy in abnormal uterine bleeding: Redefining diagnostic strategies. Indian J Cancer [Epub ahead of print] [cited 2022 May 22]. Available from: https://www.indianjcancer.com/preprintarticle.asp?id=311633
| » Introduction|| |
Abnormal uterine bleeding (AUB) is the cause for most of the gynecological visits in all age groups. The term AUB includes a broad spectrum of menstrual abnormalities such as menorrhagia, metrorrhagia, polymenorrhoea, and postmenopausal bleeding. According to International Federation of Gynecology and Obstetrics (FIGO) classification, causes of AUB can be divided into two broad groups, structural and non-structural, i.e., PALM-COEIN (polyp; adenomyosis; leiomyoma; malignancy and hyperplasia; coagulopathy; ovulatory dysfunction; endometrial; iatrogenic; and not yet classified). Amongst these, maximum cases are contributed by the PALM group, especially in late-reproductive, perimenopausal and menopausal women.
In developed countries, endometrial cancer (EC) is the most common genital tract malignancy in women. Although, only 10% of postmenopausal bleeding is due to EC, and just 10% of EC occurs before menopause, many women undergo endometrial biopsy ascribed to cancer phobia. On the other hand, many endometrial polyps and atypical hyperplasias are missed owing to non-targeted endometrial biopsy. Although, hysteroscopy overcomes these drawbacks of endometrial curettage but is costly, invasive, and cannot visualize beyond the endometrial cavity. Besides, hysteroscopy may require anesthesia, mainly postmenopausal bleeding. Transvaginal sonography (TVS) with power Doppler (PD) is a promising tool in this regard.
PD has been utilized in assessing endometrium and myometrium for about a decade. This technique can depict the microvasculature patterns in the sub-endometrial zone. PD assesses the vasculature of a lesion based on the amplitude of the Doppler signal unlike the Doppler frequency shift used in color Doppler. It is independent of the insonation angle and is perceptive of low-velocity blood vessels, depicting vasculature clearly and reliably. It is especially useful where the color Doppler signals are weak because blood vessels are small, e.g. the endometrium.,, There was a lack of standard TVS and PD features for diagnosing EC, until the introduction of the International Endometrial Tumor Analysis (IETA) group consensus nomenclature in 2010. Recent studies,,, also support using PD along with morphological features of the endometrium for suspected endometrial malignancy and premalignant conditions.
Still, there is insufficient data comparing PD sonography (PDS) and hysteroscopy in AUB cases. In this study, we aimed to assess whether TVS with PD is as accurate as hysteroscopy with guided biopsy in diagnosing EC, endometrial hyperplasia, and endometrial polyp. We have also built diagnostic protocol in AUB cases in different age groups.
| » Material and Methods|| |
A prospective cohort study was conducted in the Department of Obstetrics and Gynaecology, Institute of Medical Sciences, Banaras Hindu University from July 2012 to July 2014. A total of 142 patients with AUB were included in the study. Twenty-three women were excluded from the study because of the absence of color signals from the endometrium. Nineteen patients did not turn up for follow-up; hence,100 patients were part of the study. The study was conducted according to the guidelines of clinical studies described in the Declaration of Helsinki (as revised by the World Medical Association). Approval from the institutional ethical committee was taken. Women with AUB i.e., menorrhagia, polymenorrhoea, metrorrhagia, and postmenopausal bleeding were included. There were 45, 35, and 20 women in perimenopausal, reproductive, and postmenopausal groups, respectively. Exclusion criteria included pregnancy and related conditions, pelvic inflammatory disease, hypersensitivity to anesthetic drugs, and patients on hormonal medications. Informed consent was taken from all patients. These women were subjected to detailed history taking, complete general and gynecological examination, and routine investigations, i.e. complete blood counts, renal and liver function tests, thyroid function tests, and coagulation profile.
Firstly, a greyscale ultrasound to assess endometrial thickness and echogenicity was done. Uniformity of endometrial echogenicity, endomyometrial junction, and myometrium were assessed. PD mode was turned on. The setting conditions for this study were standardized using a transvaginal 3–9 MHz volume transducer, pulse repetition frequency at 0.6 kHz. PD parameters used were the number of blood vessels crossing the myometrial- endometrial border, size, regularity, and branching of blood vessels, areas of densely packed blood vessels, and color splashes. Vascular patterns as defined by IETA nomenclature were used, namely, multiple-vessel, irregular branching corresponding to EC [Figure 1], single-vessel with or without division as a polyp [Figure 2] and [Figure 3], and scattered-vessel pattern suggestive of hyperplasia [Figure 4], respectively. Endometritis was diagnosed if there was asymmetric dilation of the uterine cavity and asperity of endometrium's interface and zones of increased echogenicity of different sizes and forms, harboring areas of decreased echogenicity. The fibroid was identified by peripheral blood flow with the avascular center, and spokes-wheel/draping pattern. Adenomyosis was recognized by the presence of globular uterine enlargement, myometrial lakes, and obscure and thick endometrial- myometrial transformation zone. PD parameters have been proposed for polyp, hyperplasia, fibroid, and carcinoma in previous studies. For the rest of the pathologies, PD flow was turned on to see if any particular pattern could be identified.
|Figure 1: Power Doppler sonography showing multiple vessel pattern in the uterus; suggesting endometrial carcinoma|
Click here to view
|Figure 2: Power Doppler sonography showing single vessel pattern in the uterus suggesting endometrial polyp|
Click here to view
|Figure 3: Power Doppler Sonography endometrial polyp- Single feeding vessel|
Click here to view
|Figure 4: Power Doppler Sonography endometrial hyperplasia- scattered vascular pattern|
Click here to view
Hysteroscopy and guided endometrial biopsy within 24 hours of PDS were done under paracervical block or intravenous sedation or general anesthesia. Conventional panoramic hysteroscope with inner 4 mm telescope and outer 5 mm sheath was used. Proliferative endometrium was seen as pink with some superficial vascularization, secretory endometrium looked pale and orange, the polyp was seen as a localized fleshy mass with vessels on its surface, hyperplasia was noted as diffuse irregularities with increased vascularity, carcinoma showed growth with increased vascularity and areas of hemorrhage and necrosis [Figure 5]. A biopsy was taken from pathological areas. In patients with normal hysteroscopic findings, samples were taken from the fundus, anterior wall, and posterior endometrium.
|Figure 5: Hysteroscopy : Unhealthy endometrium with areas of hemorrhage : Histopathology - endometrial carcinoma|
Click here to view
Hysteroscopy was not done in three cases which suggested EC in PD [Figure 1] and [Figure 6]. We did simple dilatation and curettage in these women, taking care to obtain samples from each wall; histopathology in all three cases confirmed EC. All patients were kept in the postoperative room after hysteroscopy and observed for 4 h. In this study, none of the patients developed any life-threatening complications and all patients were managed well.
Results of both PDS and hysteroscopy were correlated and compared with the gold standard, i.e., histopathology. Data were summarized using mean and standard deviation for quantitative variables and frequency and percentage for qualitative ones. Groups were compared using paired students t-test for quantitative variables and Chi-square and Fisher's exact test for qualitative ones. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of individual parameters were calculated and compared. P value ≤ 0.05 was considered statistically significant for all tests used. All statistical analyses were performed using the SPSS 11.0 statistical package. Evaluation of the individual ability of the different PD parameters used for predicting carcinoma endometrium was plotted on “Receiver operating characteristic” (ROC) curves.
| » Results|| |
The study population consisted of 45, 35, and 20 women in perimenopausal, reproductive and postmenopausal groups, respectively. The mean age was 41.34 ± 7.67 (range 35–51) and in postmenopausal women 57.23 ± 7.52 (range 53–69 years). The most common complaint was menorrhagia [Table 1]. Three cases of carcinoma were in postmenopausal women while one in perimenopausal. All these women were obese. Hyperplasia was more common in perimenopausal women and no specific distribution for endometrial polyps was found. The demographic characteristics of the population are mentioned in [Table 2].
PD revealed EC in three (multiple irregular branching vessels), endometrial hyperplasia (scattered vessels) in nine, endometrial polyp (single vessel pattern) in nine women, rest of the findings are mentioned in [Table 3]. Here, hysteroscopy picked one case of carcinoma which was missed by PD [Figure 7]. Hysteroscopy was not done in three patients suspected to have EC by PD. They underwent dilatation and curettage as mentioned above. For carcinoma endometrium, PD had sensitivity and specificity of 75% and 100% and hysteroscopy had sensitivity and specificity of 100%, respectively. Three cases of EC were rightly picked up by PD, however, it missed two cases of the endometrial polyp and two hyperplasias. One case of hyperplasia was falsely labeled as endometritis. PD showed high sensitivity and specificity for carcinoma endometrium, hyperplasia, and endometrial polyp, which is comparable to hysteroscopy. The best vascular predictor for carcinoma endometrium was irregular branching of blood vessels closely followed by color splashes (P < 0.05).
|Figure 7: Endometrial Carcinoma- the case missed on Power Doppler Sonography - Insufficient colour signals|
Click here to view
| » Discussion|| |
In this study, we tried to find if PD can replace hysteroscopy in AUB cases. There are very few studies, comparing PDS with hysteroscopy in such a wide age distribution. We included women over a wide range of 35–69 years, i.e. late reproductive, premenopausal, and postmenopausal. We encountered a wide variety of cases and pathologies. The mean age was 41.34 ± 7.67 years in premenopausal and reproductive women and 57.23 ± 7.52 years in postmenopausal women. The mean parity of the study subjects was 3.5 ± 1.5 in both groups. The mean endometrial thickness was 18.9 ± 8.74, 14.97 ± 4.56 and 11.77 ± 5.02 in carcinoma, hyperplasia, and polyp, respectively [Table 4]. Benign lesions such as leiomyoma, polyp, and adenomyosis were more common in younger women (35–45 years). With the increase in age, more women had hyperplasia, disordered proliferative endometrium, cystic atrophy, and carcinoma.
|Table 4: Endometrial thickness in various cases of abnormal uterine bleeding|
Click here to view
We noticed that PDS is comparable with hysteroscopy in endometrial cavity lesions (P < 0.05) [Table 5] and [Table 6] and superior for other myometrial lesions. For carcinoma endometrium, sensitivity, specificity, PPV, and NPV of PD were 75%, 100%, 100%, and 98.97%, respectively, in our study. This is quite similar to results by Mahmoud et al., i.e. 85.71% sensitivity, 97.26% specificity, 75% PPV, and 98.61% NPV. The values were also comparable for endometrial hyperplasia and polyp [Table 5]. Kurjak et al. and Kupesic et al. proved the utility of morphologic criteria and 3D PD findings in EC; with sensitivity and specificity of 89 and 97%, respectively. Kabil Kucur et al. elucidated that single dominant artery with or without branching strongly suggested endometrial polyps (P = 0.001); and multiple vessels with focal origin pattern significantly correlated with EC (P < 0.026), a circular peripheral flow is exclusive to submucous fibroids (P = 0.001), and scattered vessel pattern was suggestive of endometrial hyperplasia (P = 0.001). Szpurek et al. reported the presence of “new, irregular vessels” in 81% of women with EC and 12% of women with hyperplasia. However, according to a few authors, there is no specific vascular pattern with a statistically significant association with EC.
Opoloskine et al. suggested that the PD variable that best predicted the malignancy was irregular branching of endometrial blood vessels. We also interpreted irregular branching of endometrial blood vessels, closely followed by color splashes as the best parameters for the prediction of EC (P < 0.05). The scope of PD can be estimated by the fact that some studies propose that it can predict tumor invasiveness and lymph node invasion, another advantage over the greyscale ultrasound and hysteroscopy.,,,, Blood flow parameters of the uterine arteries may also be useful in differentiating endometrial lesions.,, Increasing color score can predict increasing tumor grade, stage, infiltration, nodal metastasis, and lymph vascular space invasion. The abovementioned parameters could form the basis of the next large-scale study for the prediction of EC. Besides, with the advent of molecular classification of EC,, it will be fascinating if PD could predict the genome of EC and help in the personalized management of these patients.
A possible limitation of our study is the size of the study population. The results may vary if a bigger population is taken. To make PD findings reproducible and to standardize the findings, IETA consensus nomenclature needs to be followed and regular training of radiologists and gynecologists to be ensured. More ultrasound scores could be developed for the prediction of premalignant and malignant endometrium. Another possible limitation is the unavailability of the endometrial sample if hysteroscopy is omitted. However, this study does not advocate the omission of endometrial sampling in suspicious malignant and premalignant cases. We put forward PD TVS in AUB as a primary screening tool for deciding further tests [Figure 8] and [Figure 9]. In other words, if there is a high certainty of malignancy, an endometrial biopsy is advocated, but hysteroscopy is resisted due to the risk of possible spillage of tumor cells retrogradely.
|Figure 8: Diagnostic protocol AUB - Reproductive and peri- menopausal age. AUB: Abnormal uterine bleeding, TVS: Transvaginal sonography, ET: Endometrial thickness, PD: Power Doppler, Dx: Diagnosis |
Click here to view
|Figure 9: Diagnostic protocol: Post- menopausal Bleeding. TVS: Transvaginal sonography, ET: Endometrial thickness, Dx: Diagnosis|
Click here to view
Our study has thoroughly compared PDS with hysteroscopy, a feature exclusive from prior studies on a similar topic. The differences in results between our study and previous studies can almost certainly be explained by substantial differences in study populations and study design. There are differences in menopausal status, use of hormone replacement therapy, and racial difference.
| » Conclusion|| |
PD ultrasound is an underutilized modality which can be used to supplement hysteroscopy for screening purpose. Irregular branching vessels and color splashes are most predictive of malignancy; hence the presence of these findings should be followed by thorough endometrial curettage. Hysteroscopy should be restricted in such cases. Women with scattered blood vessels, indicating hyperplasia and those with single pedicle vessels should undergo therapeutic hysteroscopy. When there is peripheral circular flow suggesting fibroid, a further decision should be made according to age, site and size of fibroid, and desire for fertility.
Standardization of nomenclature, as suggested by IETA clinicians, between sonographers should be encouraged. Large-scale studies on this technology to explore all aspects of EC such as tumor stage, grade, lymphovascular space invasion, and molecular classification should be attempted. We propose to include PDS in all TVS cases of AUB for a better preoperative understanding of the disease. Since the sample size in the present study was small, studies with larger sample size are required to consolidate the findings of the present study.
Declaration of patient consent
The authors certify that they have obtained all appropriate patient consent forms. In the form, the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.
We thank all the patients who took part in the study.
We also thank all the teachers for guiding us through the study.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| » References|| |
Munro MG, Critchley HO, Broder MS, Fraser IS. FIGO classification system (PALM-COEIN) for causes of abnormal uterine bleeding in nongravid women of reproductive age. Int. J. Gynecol Obstet 2011;113:3-13.
Alcazar JL, Castillo G, Minguez JA, Galan MJ. Endometrial blood flow mapping using transvaginal power Doppler sonography in women with postmenopausal bleeding and thickened endometrium. Ultrasound Obstet Gynecol 2003;21:583-8.
Breijer MC, van Doorn HC, Clark TJ, Khan KS, Timmermans, A, Mol BW, et al
. Diagnostic strategies for endometrial cancer in women with postmenopausal bleeding: Cost-effectiveness of individualized strategies. Eur J Obstet Gynecol Reprod Biol 2012;163:91-6.
Van Hanegema N, Breijer MC, Khan KS, Clark TJ, Burger MP, Mol BW, et al
. Diagnostic evaluation of the endometrium in postmenopausal bleeding. Maturitas 2011;68:155-64.
Gawron I, Loboda M, Babczyk D, Ludwin I, Basta P, Pitynski K, et al
. Endometrial cancer and hyperplasia rate in women before menopause with abnormal uterine bleeding undergoing endometrial sampling. Przegl Lek 2017;74:139-43.
Kabil Kucur S, Temizkan O, Atis A, Gozukara I, Uludag EU, Agar S, et al
. Role of endometrial power doppler ultrasound using the international endometrial tumor analysis group classification in predicting intrauterine pathology. Arch Gynecol Obstet 2013;288:649-54.
Martinoli C, Derchi LE, Rizatto G, Solbiati L. Power doppler sonography: General principles, clinical applications, and future prospects. Eur Radiol 1998;8:1224-35.
Idowu BM, Ibitoye BO. Doppler sonography of perifibroid and intrafibroid arteries of uterine leiomyomas. Obstet Gynecol Sci 2018;61:395-403.
Leone FP, Timmerman D, Bourne T, Valentin L, Epstein E, Goldstein SR, et al
. Terms, definitions and measurements to describe the sonographic features of the endometrium and intrauterine lesions: A consensus opinion from the international endometrial tumor analysis (IETA) group. Ultrasound Obstet Gynecol 2010;35:103-12.
Epstein E, Fischerova D, Valentin L, Testa AC, Franchi D, Opoloskine G, et al
. Ultrasound characteristics of endometrial cancer as defined by International Endometrial Tumor Analysis (IETA) consensus nomenclature: Prospective multicenter study. Ultrasound Obstet Gynecol 2018;51:818-28.
Colombo N, Creutzberg C, Amant F, Bosse T, Gonzalez-Martin A, Ledermann J, et al
. ESMO-ESGO-ESTRO Consensus conference on endometrial cancer: Diagnosis, treatment, and follow-up. Ann Oncol 2016;27: 16-41.
Epstein E, Van Holsbeke C, Mascilini F, Masback A, Kannisto P, Ameye L, et al
. Gray-scale and color doppler ultrasound characteristics of endometrial cancer in relation to stage, grade and tumor size. Ultrasound Obstet Gynecol 2011;38:586-93.
Arslan M, Erdem A, Erdem M, Yazici G, Himmetoglu O, Gursoy R. Transvaginal color doppler ultrasonography for prediction of pre-cancerous endometrial lesions. Int. J. Gynaecol Obstet 2003;80:299-306.
Pyari JS, Rekha S, Srivastava PK, Goel M, Pandey M. A comparative diagnostic evaluation of hysteroscopy, transvaginal ultrasonography and histopathological examination in cases of abnormal uterine bleeding. J Obstet Gynecol India 2006;56:240-3.
Mahmoud EM, Mohammed HM, Ahmed FG, Ahmed NM. Transvaginal power doppler sonography can discriminate between benign and malignant endometrial conditions in women with postmenopausal bleeding. Middle East Fertil Soc J 2012;17:22-9.
Kurjak A, Kupesic S, Sparac V, Bekavac I. Preoperative evaluation of pelvic tumors by doppler and three-dimensional sonography. J Ultrasound Med 2001;20:829-40.
Kupesic S, Kurjak A, Hajder E. Ultrasonic assessment of the postmenopausal uterus. Maturitas 2002;41:255-67.
Szpurek D, Sajdak S, Moszynski R, Roszak A. Estimation of neovascularisation in hyperplasia and carcinoma of endometrium using a 'power' angio-doppler technique. Eur J Gynaecol Oncol 2000;21:405-7.
Opolskiene G, Sladkevicius P and Valentin L. Ultrasound assessment of endometrial morphology and vascularity to predict endometrial malignancy in women with postmenopausal bleeding and sonographic endometrial thickness >or = 4.5 mm. Ultrasound Obstet Gynecol 2007;30:332-40.
Yasser I, Abdelkhalek, Mohamed GM, Farouk O. Three-dimensional transvaginal sonography and power doppler angiography in the differentiation between endometrial hyperplasia and endometrial carcinoma in postmenopausal women with abnormal uterine bleeding. Egypt J Radiol Nucl Med 2016;47:1795-1801.
Galván R, Mercé L, Jurado M, Mínguez JA, López-García G, Alcázar JL. Three-dimensional power doppler angiography in endometrial cancer: Correlation with tumor characteristics. Ultrasound Obstet Gynecol 2010;35:723-9.
Saarelainen SK, Vuento MH, Kirkinen P, Mäenpää JU. Preoperative assessment of endometrial carcinoma by three-dimensional power doppler angiography. Ultrasound Obstet Gynecol 2012;39:466-72.
El-Sharkawy M, El-Mazny A, Ramadan W. Three-dimensional ultrasonography and power doppler for discrimination between benign and malignant endometrium in premenopausal women with abnormal uterine bleeding. BMC Women's Health 2016;16:18.
Kim A, Lee JY, Chun S, Kim HY. Diagnostic utility of three-dimensional power Doppler ultrasound for postmenopausal bleeding. Taiwan J Obstet Gynecol 2015;54:221-6.
Bezircioglu I, Baloglu A, Cetinkaya B, Yigit S, Oziz E. The diagnostic value of the Doppler ultrasonography in distinguishing the endometrial malignancies in women with postmenopausal bleeding. Arch Gynecol Obstet 2012;285:1369-74.
Wilczak M, Samulak D, Englert-Golon M, Pieta B. Clinical usefulness of evaluation of quality parameters of blood flow: Pulsation index and resistance index in the uterine arteries in the initial differential diagnostics of pathology within the endometrium. Eur J Gynaecol Oncol 2010;31:437-9.
Talhouk A, McAlpine JN. New classification of endometrial cancers: The development and potential applications of genomic-based classification in research and clinical care. Gynecol Oncol Res Pract 2016;3:14.
Kandoth C, Schultz N, Cherniack AD, Akbani R, Liu Y, Shen H, Robertson AG, Pashtan I et al
. Integrated genomic characterization of endometrial carcinoma. Nature 2013;497:67-73.
Madkour NM. An ultrasound risk-scoring model for prediction of endometrial cancer in post-menopausal women (using IETA terminology). Middle East Fertil Soc J 2017;22:201-5.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7], [Figure 8], [Figure 9]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6]