SEARCH

SEARCH BY CITATION

Keywords:

  • cancer;
  • plasma fibrinogen;
  • prognosis;
  • urothelial carcinoma;
  • upper tract

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Aknowledgements
  8. Conflict of Interest
  9. References

What's known on the subject? and What does the study add?

  • Upper tract urothelial carcinoma (UTUC) is relatively uncommon, accounting for only ∼5% of urothelial malignancies and 10% of all renal tumours. Radical nephroureterectomy (RNU) with bladder cuff excision is the surgical standard of care for treating localized UTUC, but the prognosis for patients who undergo RNU remains poor. Evidence suggests that an interactive relationship exists between haemostatic factors and tumour biology. A number of procoagulant and fibrinolytic factors have been found to be overexpressed in tumours. One of these factors is plasma fibrinogen. Recent studies have shown that elevated pre-therapeutic plasma fibrinogen levels are associated with worse outcome in various malignancies; however, the prognostic value of plasma fibrinogen levels for UTUC has not yet been reported.
  • To the best of our knowledge, this is the first paper to evaluate the prognostic impact of preoperative plasma fibrinogen levels in patients with localized UTUC treated surgically. We believe that the present results may assist in decision-making with respect to the need for lymph node dissection and neoadjuvant chemotherapy.

Objective

  • To investigate the prognostic value of plasma fibrinogen levels as a predictor of patient outcome in upper tract urothelial carcinoma (UTUC).

Patients and Methods

  • A total of 218 patients who underwent radical nephroureterectomy (RNU) for localized UTUC (pTa-4N0M0) were identified between 1995 and 2009.
  • The association between preoperative plasma fibrinogen levels and clinicopathological variables was analysed.

Results

  • Forty-five patients experienced tumour recurrence, and 36 died from disease during the mean follow-up of 51 months. The mean (sd) preoperative plasma fibrinogen level was 362 (103) mg/dL.
  • Kaplan–Meier curves showed that subsequent tumour recurrence was strongly predicted in patients with preoperative plasma fibrinogen levels ≥450 mg/dL, and similar results were observed for cancer-specific survival.
  • On multivariate analysis we found that a preoperative plasma fibrinogen level of ≥450 mg/dL was an independent risk factor for subsequent tumour recurrence and cancer-specific survival.
  • The 5-year recurrence-free survival rate was 56.9% in patients with plasma fibrinogen levels ≥450 mg/dL and 81.5% in patients with plasma fibrinogen levels <450 mg/dL (P < 0.001). The 5-year cancer-specific survival rate was 59.5% in patients with plasma fibrinogen levels of ≥450 mg/dL and 84.8% in patients with plasma fibrinogen levels <450 mg/dL (P < 0.001).
  • On multivariate analysis, controlling for preoperative indicators, a preoperative plasma fibrinogen level of ≥450 mg/dL predicted worse pathological features, such as ≥pT3 disease and positive lymphovascular invasion, in surgical specimens.

Conclusions

  • Preoperative elevated plasma fibrinogen level was an independent predictor for poor survival after RNU and for worse pathological features.
  • Plasma fibrinogen levels may become a useful biomarker, particularly because of its low associated cost and easy accessibility.

Abbreviations
UTUC

upper tract urothelial carcinoma

RNU

radical nephroureterectomy

LVI

Lymphovascular invasion

CIS

carcinoma in situ

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Aknowledgements
  8. Conflict of Interest
  9. References

Upper tract urothelial carcinoma (UTUC) is relatively uncommon, accounting for only ∼5% of urothelial malignancies and 10% of all renal tumours [1]. Although radical nephroureterectomy (RNU) with bladder cuff excision is the surgical standard of care for treating localized UTUC, the prognosis for patients who undergo RNU still remains poor because of the risk of local or distant recurrent disease. Many patients with local UTUC have ≥ T3 stage disease at the time of surgery [2], and up to 30% of patients with UTUC already have lymph node involvement at diagnosis [3]. Even after curative therapy, 5-year disease-specific survivals ranging between 50% and 80% have been reported [4-6].

A growing body of evidence suggests that an interactive relationship exists between haemostatic factors and tumour biology, in terms of their influence on cancer development, growth and metastasis [7]. A number of procoagulant and fibrinolytic factors have been found to be overexpressed in tumours [8, 9]. One of these factors is fibrinogen, a plasma glycoprotein that plays a key role in clot formation and wound healing and binds to platelets to support platelet aggregation, which is the final step in the coagulation cascade. Recent studies have shown that elevated pre-therapeutic plasma fibrinogen levels are associated with worse outcome in various malignancies [10, 11]; however, the prognostic value of plasma fibrinogen levels for UTUC has not yet been reported.

Given the high rate of subsequent cancer progression, including local or distant recurrence in patients treated with RNU, we decided to investigate preoperative plasma fibrinogen levels as a predictor of patient prognosis after surgery. Furthermore, we also evaluated whether elevated preoperative plasma fibrinogen levels could predict worse pathological features in surgical specimens.

Patients and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Aknowledgements
  8. Conflict of Interest
  9. References

After institutional review board approval, the medical records from 1995 to 2009 archived at Keio University Hospital and Saiseikai Central Hospital were reviewed. During this period, 251 patients underwent RNU for UTUC at these two institutions, and a total of 235 patients were identified as having localized UTUC (Ta-4N0M0) for inclusion the study. Three patients with concomitant carcinoma invading bladder muscle and two patients who underwent cisplatin-based neoadjuvant chemotherapy were excluded. After excluding patients lost to follow-up within 3 months, patients without data on preoperative plasma fibrinogen levels, patients with an active infection, a haematological disorder, or acute or chronic inflammatory and/or autoimmune disease, and those who had undergone previous steroid therapy, there were 218 remaining patients who were included in the analyses. The median (range) follow-up for the entire cohort was 38 (3–187) months.

Open RNU was performed in 155 patients (71.1%), while the remaining 63 patients (28.9%) underwent laparoscopic RNU. Dissection of regional lymph nodes was performed in patients with nodes that were found to be enlarged in a preoperative evaluation, or in those who were suspected of having enlarged nodes at intraoperative inspection. Extended lymphadenectomy was not routinely performed. Postoperative adjuvant radiotherapy regimens were not routinely used.

Surgical specimens were processed according to standard pathological procedure at each institution. All specimens were histologically confirmed to be urothelial carcinoma. Tumours were staged according to the 2002 American Joint Committee on Cancer/Union International Contre le Cancer TNM classification. Tumour grade was assessed according to the WHO classification of 1973. Lymphovascular invasion (LVI) was defined as the presence of tumour cells within an endothelium-lined space without underlying muscular walls. The presence of concomitant carcinoma in situ (CIS) was also assessed in every representative section. Blood samples were collected before any surgical intervention. Blood data concerning plasma fibrinogen levels were determined using the Clauss method [12] in a consecutive fashion during the study period. Preoperative plasma fibrinogen levels of 180–390 mg/dL were defined as normal [11].

Patients were assessed by urine cytology and cystoscopy every 3 months for 2 years after RNU, every 6 months for the next 3 years, and every 6–12 months thereafter. CT, MRI and/or excretory urograms were also performed every 6 months for 5 years and annually thereafter. Disease recurrence was defined as any documented recurrence by radiograph or pathology-proven failure in non-bladder lesions such as contralateral kidney, operative site, regional lymph nodes, or distant metastasis. The cause of death was determined by the attending physicians at each institution.

Statistical Analysis

Values are reported as means (sd). The variables of different groups were compared using the chi-squared test, Mann–Whitney U-test, or Kruskal-Wallis test, as appropriate. Recurrence-free and cancer-specific survival were estimated using the Kaplan–Meier method. Survival time of patients who were disease-free or still alive or who had died from other causes was censored at the last follow-up data, and survival curves were compared using the log-rank test. Univariate and multivariate analysis for recurrence-free and cancer-specific survival was carried out using Cox proportional hazards regression models with stepwise forward selection. Preoperative variables, including patient age, gender, clinical T stage, urine cytology, tumour length, tumour location and preoperative plasma fibrinogen levels, and postoperative variables, including tumour grade, pathological T stage, the appearance of LVI, the presence of concomitant CIS and the status of adjuvant chemotherapy, were used in survival analysis. Clinical T stage, tumour length, and tumour location were determined by CT and/or MRI preoperatively. Tumour location was divided into two areas: the renal pelvis or ureter, based upon the location of the dominant lesion.

Logistic regression methodology was used to predict higher pathological T stage and positive LVI in surgical specimens. Preoperative variables, including patient age, gender, clinical T stage, urine cytology, tumour length, tumour location, and preoperative plasma fibrinogen levels were used in logistic regression analyses. Differences among groups were regarded as significant when P < 0.05. These analyses were performed using the SPSS version 17.0 statistical software package.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Aknowledgements
  8. Conflict of Interest
  9. References

The mean (range) age of all cohorts was 69 (38–92) years. Males accounted for 73.4% (160 patients) and females 26.6% (58 patients). Table 1 shows the clinicopathological variables in the 218 patients. Cisplatin-based adjuvant chemotherapy regimens were administered after RNU to 42 patients (19.3%). The mean (sd) preoperative plasma fibrinogen level in the whole cohort was 362 (103) mg/dL. Preoperative plasma fibrinogen levels were significantly associated with tumour length, tumour grade, pathological T stage, and positive LVI in surgical specimens, and the status of adjuvant chemotherapy.

Table 1. Patient characteristics and plasma fibrinogen levels in the 218 study patients
CharacteristicNo. of patients (%)Mean (sd) plasma fibrinogen levels, mg/dLP
Preoperative variables   
Age  0.133
<70 years93 (42.7)354 (106)
≥70 years125 (57.3)368 (100)
Gender  0.669
Male160 (73.4)364 (108)
Female58 (26.6)355 (85)
Clinical T stage  0.280
< cT3126 (57.8)358 (104)
≥ cT392 (42.2)368 (101)
Urine cytology  0.576
Positive61 (28.0)379 (132)
Negative157 (72.0)356 (88)
Tumour length  0.022
<30 mm126 (57.8)350 (103)
≥30 mm92 (42.2)378 (101)
Tumour location  0.344
Renal pelvis130 (59.6)358 (106)
Ureter88 (40.4)366 (97)
Postoperative variables   
Tumour grade  <0.001
G1/259 (39.6)326 (80)
G3159 (60.4)375 (107)
Pathological T stage  0.007
pTa-175 (34.4)345 (85)
pT227 (12.4)337 (83)
pT3107 (49.1)372 (111)
pT49 (4.1)468 (114)
LVI  0.001
Positive84 (38.5)393 (116)
Negative134 (61.5)342 (88)
Concomitant CIS  0.331
Positive53 (24.3)368 (97)
Negative165 (75.7)360 (105)
Adjuvant chemotherapy  0.002
Yes42 (19.3)409 (127)
No176 (80.7)351 (93)

During the mean follow-up of 38 months, 45 (21.6%) of 218 patients experienced tumour recurrence. Figure 1 shows the Kaplan–Meier curve of recurrence-free survival according to various plasma fibrinogen level thresholds. Patients with plasma fibrinogen levels ≥450 mg/dL could be predicted to have a higher risk of subsequent tumour recurrence (Fig. 1A; P < 0.001), while similar results were observed by using other thresholds such as a plasma fibrinogen level of ≥420 (Fig. 1B) or ≥390 mg/dL (Fig. 1C). The 5-year recurrence-free survival rate was 56.9% in patients with plasma fibrinogen levels ≥450 mg/dL and 81.5% in patients with plasma fibrinogen levels <450 mg/dL (Fig. 1A), 63.6% in patients with plasma fibrinogen levels ≥420 mg/dL and 82.0% in patients with plasma fibrinogen levels <450 mg/dL (Fig. 1B), and 66.5% in patients with plasma fibrinogen levels ≥390 mg/dL and 82.9% in patients with plasma fibrinogen levels <450 mg/dL (Fig. 1C).

figure

Figure 1. Recurrence-free survival rate after RNU for a plasma fibrinogen level (A) <450 vs ≥450, (B) <420 vs ≥420, or (C) <390 vs ≥390 mg/dL.

Download figure to PowerPoint

Using the defined parameter of a plasma fibrinogen level of ≥450 mg/dL, we examined the prognostic impact of preoperative plasma fibrinogen level on subsequent tumour recurrence. The association between preoperative plasma fibrinogen levels and the clinicopathological variables of the patient are shown in Table 2. Patients with preoperative plasma fibrinogen levels ≥450 mg/dL had higher incidences of positive urine cytology at diagnosis, tumour grade 3, ≥ pT3 stage, and positive LVI in surgical specimens, and tended to have undergone adjuvant chemotherapy; however, no significant difference was found among patient age, gender, clinical T stage, tumour length, tumour location and the presence of CIS in surgical specimens. Univariate and multivariate analyses were performed to determine the predictors of subsequent tumour recurrence (Table 3). Multivariate analysis showed that ≥ pathological T3 stage, positive LVI, and preoperative plasma fibrinogen level of ≥450 mg/dL were independent risk factors for subsequent tumour recurrence.

Table 2. Clinicopathological variables in 218 patients according to plasma fibrinogen levels
CharacteristicPatients with plasma fibrinogen <450 mg/dL (%)Patients with plasma fibrinogen ≥450 mg/dL (%)P
No of patients18632 
Preoperative variables   
Age  0.521
<70 years79 (42.5)14 (43.7)
≥70 years107 (57.5)18 (56.3)
Gender  0.114
Male134 (72.0)26 (81.3)
Female52 (28.0)6 (18.7)
Clinical T stage  0.502
<cT3107 (57.5)19 (59.4)
≥cT379 (42.5)13 (40.6)
Urine cytology  0.029
Positive47 (25.3)14 (43.8)
Negative139 (74.7)18 (56.2)
Tumour length  0.061
<30 mm112 (60.2)14 (43.8)
≥30 mm74 (39.8)18 (56.2)
Tumour location  0.532
Renal pelvis112 (60.2)19 (59.4)
Ureter74 (39.8)13 (40.6)
Postoperative variables   
Tumour grade  <0.001
G1/258 (31.2)1 (3.1)
G3128 (68.8)31 (96.9)
Pathological T stage  0.008
pTa-169 (37.1)6 (18.7)
pT225 (13.4)2 (6.3)
pT387 (46.8)20 (62.5)
pT45 (2.7)4 (12.5)
LVI  0.008
Positive65 (34.9)19 (59.4)
Negative121 (65.1)13 (40.6)
Concomitant CIS  0.291
Positive47 (25.3)6 (18.7)
Negative139 (74.7)26 (81.3)
Adjuvant Chemotherapy  0.002
Yes29 (15.6)13 (40.6)
No157 (84.4)19 (59.4)
Tumour metastasis32 (17.2)13 (40.6)0.004
Deaths from disease24 (12.9)12 (37.5)0.002
Table 3. Risk factors for predicting tumour recurrence and cancer-specific survival after RNU in 218 patients
CharacteristicRecurrence-free survivalCancer-specific survival
UnivariateMultivariateUnivariateMultivariate
PHR (95% CI)PPHR (95% CI)P
Preoperative variables      
Age (<70 vs ≥70 years)0.037  0.047  
Gender (male vs female)0.049  0.100  
Clinical T stage (<cT3 vs ≥cT3)0.820  0.980  
Urine cytology (positive vs negative)0.187  0.584  
Tumour length (<30 vs ≥30 mm)0.930  0.967  
Tumour location (renal pelvis vs ureter)0.281  0.448  
Plasma fibrinogen level (<450 vs ≥450 mg/dL)<0.0012.00 (1.04–3.85)0.038<0.0012.41 (1.20–4.85)0.028
Postoperative variables      
Tumour grade (G1/2 vs G3)0.001  0.001  
Pathological T stage (<pT3 vs ≥pT3)<0.0013.64 (1.34–9.86)0.011<0.0013.43 (1.15–10.2)0.025
LVI (positive vs negative)<0.0012.43 (1.13–5.19)0.022<0.0012.19 (0.96–4.98)0.049
Concomitant CIS (positive vs negative)0.407  0.416  
Adjuvant chemotherapy (yes vs no)0.008  0.117  

A total of 36 subjects (16.5%) died from disease during follow-up. Univariate and multivariate analyses were performed to investigate whether preoperative plasma fibrinogen levels could be a biomarker for predicting cancer-specific survival. Multivariate analysis showed that ≥ pathological T3 stage, positive LVI, and preoperative plasma fibrinogen level of ≥450 mg/dL were independent risk factors for cancer-specific survival. The 5-year cancer-specific survival rate was 59.5% in patients with a plasma fibrinogen level of ≥450 mg/dL and 84.8% in patients with plasma fibrinogen levels <450 mg/dL (Fig. 2; P < 0.001). Using other pT stage thresholds, such as ≥ pT2 stage, we also examined whether plasma fibrinogen levels could also predict subsequent tumour recurrence and worse cancer-specific survival. Similarly, multivariate analysis showed that a preoperative plasma fibrinogen level of ≥450 mg/dL was an independent risk factor for decreasing subsequent recurrence-free survival and cancer-specific survival.

figure

Figure 2. Cancer-specific survival rate after RNU for a plasma fibrinogen level <450 vs ≥450 mg/dL.

Download figure to PowerPoint

Next, we evaluated the clinical impact of preoperative plasma fibrinogen levels on the prediction of worse pathological outcomes in surgical specimens, such as higher pT stage and positive LVI (Table 4). On multivariate analysis controlling for preoperative indicators, preoperative fibrinogen levels ≥450 mg/dL could be an independent and the strongest indicator for predicting ≥ pathological T3 stage and positive LVI in surgical specimens, showing the highest hazard ratio compared with other preoperative variables for each pathological feature.

Table 4. Univariate and multivariate analysis of preoperative variables prognostic for pathological T stage and LVI
 UnivariateMultivariate
PHR (95% CI)P
Pathological T3 or greater   
Age (<70 vs ≥70 years)0.339  
Gender (male vs female)0.321  
Clinical T stage (<cT3 vs ≥cT3)0.0012.51 (1.41–4.47)0.002
Urine cytology (positive vs negative)0.171  
Tumour length (<30 vs ≥30 mm)0.028  
Tumour location (renal pelvis vs ureter)0.0220.56 (0.31–0.99)0.045
Plasma fibrinogen level (<450 vs ≥450 mg/dL)0.0073.56 (1.47–8.60)0.005
Positive LVI   
Age (<70 vs ≥70 years)0.281  
Gender (male vs female)0.277  
Clinical T stage (<cT3 vs ≥cT3)0.0172.02 (1.15–3.56)0.015
Urine cytology (positive vs negative)0.045  
Tumour length (<30 vs ≥30 mm)0.473  
Tumour location (renal pelvis vs ureter)0.482  
Plasma fibrinogen level (<450 vs ≥450 mg/dL)0.0062.99 (1.38–6.50)0.005

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Aknowledgements
  8. Conflict of Interest
  9. References

In the present study, we investigated plasma fibrinogen levels and other standard prognostic factors in 218 patients undergoing RNU for localized UTUC. Kaplan–Meier analysis showed that patients with high plasma fibrinogen levels could be predicted to have a higher risk of subsequent tumour recurrence and poor cancer-specific survival. Multivariate analysis showed that in addition to other standard prognostic factors, preoperative plasma fibrinogen levels were an independent predictor of subsequent tumour recurrence as well as cancer-specific survival. Furthermore, elevated preoperative plasma fibrinogen levels independently predicted higher pathological T stage and positive LVI in surgical specimens. These data suggest plasma fibrinogen levels are a strong biomarker for predicting oncological outcome preoperatively in localized UTUC.

Coagulation pathways are often activated in cancer. It is increasingly recognized that haemostatic variables, particularly plasma fibrinogen levels, have prognostic significance in patients with cancer. Elevated plasma fibrinogen levels might be caused by the state of hyperfibrinogenaemia and hypoxia induced by tumour growth [13, 14], by the endogenous production of fibrinogen by tumour cells themselves [15], or by inflammation-mediated cells such as host macrophages and epithelial cells [16]. While fibrinogen actively exhibits a role in tumour growth, invasion and metastasis by promoting tumour neovascularization and by supporting the sustained adhesion of tumour cells [17-19], the state of pre-therapeutic plasma fibrinogen may reflect the progressive and metastatic potential of tumours. In the present study, elevated preoperative plasma fibrinogen levels were significantly associated with reported prognostic indicators, such as pathological T stage and the presence of LVI. This was in accordance with previously published data in other malignancies [10, 11].

Despite the poor patient prognosis in UTUC, many researchers have been investigating prognostic indicators to stratify the risk profiles for this disease. Large multi-institutional studies have also attempted to identify possible prognostic indicators according to a variety of patient characteristics [4, 20, 21]. As a result, the accuracy of survival prediction in UTUC has been improved. Such studies would be more impressive if they enabled us to undertake additional interventions in patients who were most likely to benefit. While these studies assessed indicators which were not available before RNU, possible intervention based on the analyses might be limited to the administration of adjuvant chemotherapy for high risk tumours; however, no significant data suggesting the benefit of adjuvant chemotherapy are currently available [1, 22, 23], and other alternatives need to be explored.

One of the possible interventions is neoadjuvant chemotherapy. For patients with UTUC, neoadjuvant chemotherapy has the advantage of a more effective delivery of chemotherapy owing to better renal function [24]. By extrapolating the experiences from bladder cancer, neoadjuvant chemotherapy has the potential to play a significant role in UTUC [25]. Similarly, lymph node dissection at nephroureterectomy may also have a therapeutic potential to improve disease outcomes [1]. However, unlike bladder cancer, it is difficult to identify pathological features accurately, such as the presence of muscle invasion and tumour grade before RNU. Although there has been some success in stage prediction based on imaging and tumour grade at biopsy using ureteroscopy, uncertainty rates are still in the order of 30% to 40% [26]. Thus, appropriate selection of the candidate for the latter two strategies can be challenging.

To increase the accuracy of preoperative prediction of advanced UTUC stage or poor survival, several investigators recently attempted to identify the potential solutions by using various imaging and biopsy features. Ipsilateral hydronephrosis, high tumour grade at biopsy, and positive urinary cytology are potentially associated with more advanced UTUC pathology at RNU [27, 28]. Brien et al. [29] reported that the combination of these tests could improve the prediction of advanced stage UTUC, but previous studies did not fully evaluate the prognostic roles of plasma markers concerning UTUC outcomes, even though these could be easily obtained before RNU.

To the best of our knowledge, this is the first paper to evaluate the prognostic impact of preoperative plasma fibrinogen levels in patients with localized UTUC treated surgically. We believe that the present results may assist in decision-making with respect to the need for lymph node dissection and neoadjuvant chemotherapy; however, although our study focused only on plasma fibrinogen levels, we did not fully evaluate the associations between plasma fibrinogen and other biomarkers. This may be a limitation of our study, and future validation would be needed.

The present study has some further limitations. First, it was performed in a retrospective manner and in a limited number of patients, thus unknown sources of bias may exist in the findings. Also, not all patients received adjuvant chemotherapy, which may have had an effect on subsequent patient outcomes. Prospective studies with a larger population are warranted so as to determine the accurate prognostic role of plasma fibrinogen levels in patients with UTUC.

In conclusion, preoperative plasma fibrinogen level was an independent predictor of patient survival after RNU in localized UTUC. Patients with an elevated preoperative plasma fibrinogen level could be predicted to have a higher risk of poor mortality and survival after surgery. Furthermore, elevated preoperative plasma fibrinogen levels independently predicted worse pathological features in surgical specimens. We propose that plasma fibrinogen levels may become a useful biomarker, particularly because of the low associated cost and easy accessibility.

Aknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Aknowledgements
  8. Conflict of Interest
  9. References

This work was supported in part by Grants-in-Aid for Scientific Research (#22791495 to Tanaka N. and #23592349 to Kikuchi E.) from the Ministry of Education, Culture, Sports, Science, and Technology of Japan. This work was also supported in part by a Keio University Grant-in-Aid for Encouragement of Young Medical Scientists (#002-002-0005 to Tanaka N) from Keio University School of Medicine.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Aknowledgements
  8. Conflict of Interest
  9. References