SEARCH

SEARCH BY CITATION

Keywords:

  • Cardiovascular disease;
  • Conventional hemodialysis;
  • Endothelium dysfunction;
  • Flow-mediated dilation;
  • In-center nocturnal hemodialysis

Abstract

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONCLUSION
  7. Acknowledgments:
  8. REFERENCES

Cardiovascular disease is the leading cause of morbidity and mortality in patients with end-stage renal disease who undergo hemodialysis and endothelial dysfunction is an early key step in the development of atherosclerosis. The aim of this study was to investigate the effect of thrice-weekly in-center nocturnal hemodialysis (INHD, 8 h per session and three sessions per week) and conventional hemodialysis (CHD, 4 h per session and three sessions per week) on endothelial dysfunction in patients with end-stage renal disease. 32 INHD and 58 matched CHD patients were enrolled, baseline and 12-month measures of blood pressure (BP), serum calcium and phosphorus, serum intact PTH (iPTH) and brachial artery flow-mediated dilation (FMD) were collected and analyzed. Baseline characteristics were similar between groups except that serum phosphorus and calcium × phosphorus were higher in the INHD group. At the 12-month follow-up, there was a significant increase in FMD (6.0 ± 1.5% to 7.1 ± 1.8%, P < 0.01) in INHD patients. Multivariate analysis showed that FMD was inversely correlated with systolic BP (SBP) (β = −0.485, P < 0.01), diastolic BP (DBP) (β = −0.428, P < 0.01), iPTH (β = −0.405, P < 0.01) and serum phosphorus level (β = −0.375, P < 0.01). There was no significant change in FMD in the CHD group. Compared with CHD, INHD improves endothelial function, and control of serum phosphorus is associated with the improvement of endothelial function.

Cardiovascular disease, mostly a consequence of accelerated atherosclerosis, has been recognized as the leading cause of morbidity and mortality in patients with end-stage renal disease (ESRD) who undergo hemodialysis (1,2). Endothelial dysfunction is an early key step in the development of atherosclerosis and has consistently been observed in patients with ESRD and may contribute to the high cardiovascular morbidity and mortality in these patients (3,4). A number of factors such as arterial hypertension, chronic fluid overload, sympathetic activation and hyperparathyroidism have been implicated in impairment of the functional arterial vessel properties in patients with renal insufficiency (5–7).

Conventional hemodialysis (CHD) (4 h per session, three sessions per week) is the most widely used modality of renal replacement therapy, but a treatment with an annual mortality rate of up to 25% (8) can hardly be considered an ideal one. There is ongoing interest in modifying standard HD regimens by focusing on increasing the time and/or frequency of HD in the hope of improving the adequacy and outcome. Previous studies of short daily HD (six sessions per week with 1.5 to 3.0 h per session) have reported improved blood pressure (BP) control, regression of left ventricular hypertrophy (LVH), higher Hb levels, and decreased risk for hospitalization (9–11). Quotidian nocturnal home HD has been reported to induce regression of LVH, decrease nocturnal hypoxemia, increase vagally mediated heart rate variability during sleep, and improve phosphorus control in selected patients (12–14). These forms of HD may have advantages over conventional HD.

However, these options present feasibility challenges (15). Indeed, short daily HD decreases available dialysis time for other patients, and patients are required to spend a large fraction of their daytime in the hospital. Quotidian home nocturnal dialysis requires the development of a home dialysis program and extensive patient training.

In-center, thrice-weekly nocturnal HD (three sessions per week with 8 h per session, INHD) may be a valuable option, providing enough dialysis doses to improve outcomes in a way that can be affordable for both dialysis centers and patients (16). Previous uncontrolled studies (17–19) of INHD have demonstrated improved BP control, higher Hb levels, and decreased serum phosphorus, compared with CHD.

Recent data have showed that dysregulation of phosphate homeostasis may contribute to the endothelial dysfunction (20–23). INHD has better serum phosphorus control than CHD, and it may have a better effect on endothelial dysfunction. Using vascular ultrasound to measure flow-mediated dilation (FMD) of the brachial artery, we performed a non-randomized control study to examine the effects of INHD compared with CHD on endothelial dysfunction in the ESRD population.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONCLUSION
  7. Acknowledgments:
  8. REFERENCES

Subjects

Of 340 adult patients who were receiving maintenance HD in the dialysis center of the Second Military Medical University Changzheng Hospital (Shanghai, China), 32 patients who voluntarily consented to receive INHD were enrolled in the present study from February 2009 and followed up for 12 months. The only eligibility criterion was willingness to participate in the INHD program, and no patient was excluded. From the remaining 308 patients, 58 patients who were receiving conventional HD and comparable with the study group in terms of age, gender, causes of renal failure, comorbidity burden, vascular access, medications and time on dialysis were selected as the control. The study protocol was approved by the institutional review board, and written informed consent was obtained from all participants.

Blood pressure, anemia and renal osteopathy were treated by the nephrologists according to the requirements of the Kidney Dialysis Outcomes Quality Initiative guidelines (K/DOQI guidelines) (24). The goal of the BP control is 130/80 mmHg postdialysis. Use of antihypertensive drugs including angiotensin converting enzyme inhibitors (ACEI), angiotensin II receptor blockers (ARB), calcium antagonists and β receptor blockers was recorded. The treatment of anemia was to achieve and maintain Hb levels within the range of 110–120 g/L. Erythropoietin (EPO) use was recorded in both groups. The target value for the treatment of calcium-phosphorus metabolism disorders was calcium 8.4–9.5 mg/dL, phosphorus 3.5–5.5 mg/dL and parathyroid hormone (PTH) 150–300 pg/mL. Use of active vitamin D3 and calcium carbonate was also recorded.

Blood pressure was measured within 1 h before dialysis on the non-fistula arm in an upright sitting position after at least 10 min rest using a Moon Rabbit sphygmomanometer (Shanghai Medical Instruments, Shanghai, China) by the same operator. Two readings were recorded for each individual. The mean of the two readings was defined as BP of the subjects. Predialysis blood routine examination, metabolic profiles and intact PTH (iPTH) status were measured monthly. Predialysis brachial artery FMD was performed before and by the end of follow-up.

Hemodialysis procedures

Both CHD and INHD were performed in-center, 4 h per session and three day sessions per week for CHD, and 8 h per session and three night sessions per week from 22.00 hours to 06.00 hours for INHD. Dialysis equipment, Fresenius 4008S (Bad Homburg, Germany) machines and ASAHI APS-15U polysulfone dialyzers (Tokyo, Japan), were the same for CHD and INHD. The surface area of dialyzer used for patients on both CHD and INHD was 1.5 m2. The blood and dialysate flows for CHD were 250–280 mL/min and 500 mL/min, and for INHD were 190–220 mL/min and 300 mL/min. Anticoagulation was achieved with the use of heparin or low molecular weight heparin. Ultrafiltration set is 0–6 L. The composition of the dialysate buffer remained the same in both modalities and consisted of Na 135–143 mEq/L, K 3.0 mEq/L, Ca 1.5 mEq/L and bicarbonate 35 mEq/L.

Blood chemistry

All samples were obtained predialysis after 12 h of fasting for measurement of fasting plasma glucose, serum albumin, total cholesterol, and triglycerides. Total plasma cholesterol and triglycerides were measured by the Oxidase Method with an Abbott Aeroset Automatic Biochemistry Analyzer using reagents from Roche Diagnostics (Mannheim, Germany). Hb levels were measured with an automated analyzer (Abbott Cell-Dyn 4.000, Abbott Park, IL, USA). Serum total Ca was measured by the Cresolphtalein complexone method using Menagent Calcium 60-s kits (Menarini Diagnostics, Florence, Italy). Serum phosphorus was measured by the ammonia molybdate complex method using Menagent Phosphofix kits (Menarini Diagnostics). iPTH was measured by electrochemiluminescence immunoassay method, using Roche E170 MODULAR Immunoassay Analyzer.

Assessment of endothelial function

Endothelial function was determined by the so-called “flow-mediated dilatation” method introduced by Celermajer et al. (25), recording the dilator response of the brachial artery to increased blood flow generated during reactive hyperemia of the downstream forearm. Reactive hyperemia was induced with a pneumatic cuff placed on the forearm distal to the scanned brachial artery, and inflated to 200 mmHg pressure for 5 min. Measurements were made by a single observer using an HP Sonos 5500 with a 7.5 MHz linar transducer (Hewlett-Packard, Andover, MA, USA).The subjects remained at rest in the supine position for at least 15 min before the examination started. Each subject's non-fistula arm was comfortably immobilized in the extended position to allow consistent recording of the brachial artery 2–4 cm above the antecubital fossa. The brachial artery was scanned in a longitudinal section, magnified using a resolution box function, and gated with the R wave of an electrocardiogram (ECG). Three adjacent measurements of end-diastolic brachial artery diameter were made from single two dimensional frames. All ultrasound images were recorded on S-VHS videotape for subsequent blinded analysis. A pneumatic tourniquet was inflated to 200 mmHg with obliteration of the radial pulse. After 5 min the cuff was deflated. Flow measurements were made 60 s postdeflation. The maximum FMD diameters were calculated as the average of the three consecutive maximum diameter measurements. The FMD was then calculated as the percent change in diameter compared with baseline resting diameters. All the digital images were analyzed off-line blindly by one experienced reader. The intra- and interobserver variabilities of the FMD were determined in 20 randomly selected recordings twice by the same observer and once each by two independent observers. The intra- and interobserver variabilities of FMD were found to be 2.0 ± 0.8% and 2.4 ± 0.9%, respectively.

Statistical analyses

Numerical data are expressed as mean ± SD, distributions of each variable were examined with the Shapiro-Wilk test, P < 0.05 was considered statistically significant. The t-test was used to compare numerical data, and the χ2 test was used to compare categorical data between groups and between the baseline and 12-month follow-up. Pearson's correlation coefficient was used to investigate the relations between FMD and SBP, DBP, iPTH, serum calcium, serum phosphorus, calcium × phosphorus, and Hb in both INHD and CHD groups, multivariable linear regression analyses were performed to examine to relationship between FMD and the variables significantly related to FMD investigated by univariate analysis. Analyses were conducted using SPSS 11.5 (SPSS, Chicago, IL, USA).

RESULTS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONCLUSION
  7. Acknowledgments:
  8. REFERENCES

Baseline characteristics

There were 32 patients in INHD group and 58 patients in the CHD group. All the subjects completed the 12-month follow-up, without dropouts or deaths. Demographic and baseline clinical and laboratory characteristics were not significantly different between INHD and CHD groups, except for higher phosphorus and higher calcium × phosphorus product in the INHD group (Table 1).

Table 1. Demographic characteristics and baseline clinical and laboratory information in patients received in-center nocturnal hemodialysis (INHD) and conventional hemodialysis (CHD)
 INHD (N = 32)CHD (N = 58)P-value
  1. ACEI, angiotensin converting enzyme inhibitors; ARB, angiotensin II receptor blockers; BP, blood pressure; NS, not significant.

Age (years)43.7 ± 12.845.8 ± 9.5NS
Sex (M/F)30/252/6NS
Cause of renal disease  NS
 Glomerulonephritis21 (66%)38 (66%) 
 Hypertension9 (28%)16 (27%) 
 Polycystic kidney disease2 (6%)4 (7%) 
Vascular access  NS
 Arteriovenous fistula27 (84%)50 (86%) 
 Longterm tunneled cuffed hemodialysis catheters5 (16%)8 (14%) 
Medication use   
 ACEI20 (62.5%)38 (65.5%)NS
 ARB25 (78%)46 (79%)NS
 Calcium antagonist15 (47%)26 (45%)NS
β receptor blocker24 (75%)42 (72%)NS
Smokers (N)8 (25%)17 (29%)NS
Duration of dialysis (months)77.9 ± 65.173.5 ± 63.9NS
Clinical variables   
 Systolic BP (mmHg)139.3 ± 17.7137.3 ± 12.7NS
 Diastolic BP (mmHg)88.6 ± 11.286.1 ± 12.0NS
 Left ventricular mass index (g/m2)114.77 ± 72.73120.7 ± 56.8NS
Laboratory values   
 Serum calcium (mg/dL)8.91 ± 0.869.01 ± 0.77NS
 Serum phosphorus (mg/dL)6.81 ± 2.536.0 ± 1.720.01
 Calcium × phosphorus (mg2/dL2)60.7 ± 21.354.1 ± 14.20.05
 Hemoglobin (g/dL)11.7 ± 1.8312.0 ± 1.27NS
 Intact parathyroid hormone (pg/mL)632.3 ± 550.0582.5 ± 523.8NS
 Serum albumin (g/dL)4.03 ± 0.533.98 ± 0.62NS
 Total cholesterol (mg/dL)158 ± 37.8165 ± 32.3NS
 Triglycerides (mg/dL)218 ± 21.7229 ± 28.5NS
 Glucose (mg/dL)130 ± 72.5145 ± 65.7NS

Dialysis parameters and BP control

There was a significant increase in Kt/V (P < 0.01) and ultrafiltration (P < 0.05) and a significant decrease in predialysis SBP (P < 0.01) and DBP (P < 0.01) in the INHD group at the end of follow-up. Predialysis and postdialysis weight did not change significantly in the INHD group. There was no significant change in BP, predialysis and postdialysis weight, Kt/V and ultrafiltration at 12 months in the CHD group. The number of patients who needed antihypertensive therapy and the number of anti-hypertensives decreased significantly in the INHD group. It is noteworthy that 15 patients in this group withdrew all antihypertensive medications by the end of follow-up, while there was no significant change in the use of antihypertensive medications in the CHD group (Table 2).

Table 2. Outcomes for in-center nocturnal hemodialysis (INHD) and conventional hemodialysis (CHD) groups
 INHDCHD
BaselineAfter 12 MonthsBaselineAfter 12 Months
  1. *P < 0.05 vs baseline, **P < 0.01 vs baseline, ***P < 0.05 between-group comparison between INHD and CHD, ****P < 0.01, between-group comparison between INHD and CHD. ACEI, angiotensin converting enzyme inhibitors; ARB, angiotensin II receptor blockers; BP, blood pressure; EPO, erythropoietin; iPTH, intact parathyroid hormone.

Predialysis weight (kg)68.5 ± 7.865.3 ± 5.765.7 ± 6.164.9 ± 5.5
Postdialysis weight (kg)65.8 ± 6.763.2 ± 5.562.9 ± 6.561.8 ± 5.3
Ultrafiltration per treatment (kg)2.68 ± 2.352.09 ± 1.27*2.8 ± 2.263.1 ± 1.45***
Midweek Kt/V0.89 ± 0.251.45 ± 0.42**0.90 ± 0.371.05 ± 0.45***
Systolic BP (mmHg)139.3 ± 17.7125.3 ± 19.0**137.3 ± 12.7133.7 ± 10.5***
Diastolic BP (mmHg)88.6 ± 11.281.0 ± 12.5**86.1 ± 12.086.0 ± 9.8***
ACEI (N)205**4135****
ARB (N)122**1715***
Calcium antagonist (N)155**2624***
β receptor blocker (N)246**4238****
Anti-hypertensives number2.220.56**21.93****
Serum calcium (mg/dL)8.91 ± 0.8610.4 ± 0.589.01 ± 0.778.82 ± 0.92
Serum phosphorus (mg/dL)6.81 ± 2.534.53 ± 1.28**6.03 ± 1.725.92 ± 1.85****
Calcium × phosphorus (mg2/dL2)60.7 ± 21.347.11 ± 8.56*54.1 ± 14.252.21 ± 10.38***
Active vitamin D3 (ug/day)0.30 ± 0.170.10 ± 0.09**0.28 ± 0.180.25 ± 0.15***
Calcium carbonate (ug/day)2.7 ± 0.90.2 ± 0.3**2.4 ± 0.72.8 ± 0.8****
Serum iPTH (pg/mL)632.3 ± 550.0315.4 ± 131.6**582.5 ± 523.8482.3 ± 318.7****
Hemoglobin (g/dL)11.7 ± 1.8312.2 ± 1.1312.0 ± 1.2712.1 ± 1.32
EPO (U/week)6800 ± 3787.43533.3 ± 3440.6**6500 ± 3952.86633.3 ± 3807.9****

Mineral metabolism

At baseline, the INHD group had higher phosphorus and calcium × phosphorus product. At 12 months follow-up, there were significant decreases in serum phosphorus (P < 0.01), calcium × phosphorus product (P < 0.05) and iPTH level (P < 0.01). The use of active vitamin D3 (P < 0.01) and calcium carbonate (P < 0.01) decreased substantially, and the serum calcium did not change significantly in the INHD group. By contrast, in the CHD group, there was no significant change in serum phosphorus level, calcium × phosphorus product, serum calcium level and the use of active vitamin D3 and calcium carbonate at the end of follow-up. After 12 months, there was a significant decrease in iPTH levels (P < 0.05) in the conventional HD group (Table 2).

Regulation of hemoglobin

There was no significant change in Hb at baseline and after follow-up in the two groups. At 12 months follow-up, the use of EPO (P < 0.01) decreased substantially, and there were five patients who withdrew EPO because of Hb>140 g/L. There was no significant change in the use of EPO in the CHD group (Table 2).

Brachial artery flow-mediated dilation

In the INHD group, the baseline FMD was 6.0 ± 1.5%, at 12 months follow-up, the FMD was 7.1 ± 1.8%. In the CHD group, the baseline FMD was 5.8 ± 1.4%, at 12 months follow-up, the FMD was 6.1 ± 1.2%. There was no significant difference in baseline FMD between the two groups. At 12 months follow-up, FMD increased significantly (P < 0.01) in the INHD group, but there was no significant change in the CHD group, and the FMD was significantly higher (P < 0.01) than that in the CHD group (Fig. 1). Univariate analysis showed that FMD was inversely related to SBP (r = −0.502, P < 0.01), DBP (r = −0.453, P < 0.01), iPTH (r = −0.432, P <  0.01), serum phosphorus level (–0.398, P < 0.01), and calcium × phosphorus (r = −0.315, P < 0.05). Multivariate analysis showed that FMD was inversely correlated with SBP (β = −0.485, P < 0.01), DBP (β = −0.428, P < 0.01), iPTH (β = −0.405, P < 0.01) and serum phosphorus level (β = −0.375, P < 0.01).

image

Figure 1. Twelve-month change in flow-mediated dilation (FMD). +P < 0.01 vs baseline, P < 0.01, between-group comparison between in-center nocturnal hemodialysis (INHD) and conventional hemodialysis (CHD).

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONCLUSION
  7. Acknowledgments:
  8. REFERENCES

The main findings of this study are: first, compared with CHD, INHD is associated with an improvement in endothelial function as determined by FMD; and second, negative correlation of serum phosphate and FMD has been observed in the study.

Acute effects of HD on endothelial dysfunction are controversial. Cross et al. (26) found a single HD treatment caused rapid clearance of endothelial toxins and transiently increased FMD. Migliacci (27) showed that HD, although largely able to remove endothelium toxic substances does not ameliorate to any significant extent endothelial function in uremic patients. Lilien (28) demonstrated that a single HD procedure contributes to a detrimental effect on endothelial function and may thereby add to the cardiovascular risk of children with ESRD.

Previous studies have shown that nocturnal hemodialysis (NHD) (8 h per session, six sessions per week) has positive results on endothelial dysfunction in uremic patients. Chan et al. reported that within 1 to 2 months after ESRD, patients are converted from CHD to NHD, and endothelial vasodilation improves markedly (29). Controlled study has demonstrated that NHD improves endothelial vasodilation, the improvement in BP control by NHD may independently lead to amelioration of endothelial function, and NHD may also favorably affect a fundamental process of endothelial repair via normalizing endothelial progenitor cell number and function (30).

In our study, at 12 months follow-up, there was no significant change in FMD in the CHD group, whereas FMD increased significantly in the INHD group. At the end of follow-up, FMD in the INHD group was significant higher than that in the CHD group. The significant increase of FMD indicates the improvement of endothelial function by INHD. This is the first time the effect of INHD on endothelial dysfunction has been studied. Fathi et al. (31) examined the value of FMD in predicting cardiovascular events in a group of 444 patients with a significant risk of cardiovascular events. They demonstrated that patients with FMD of <2% had significantly more cardiovascular events than those with FMD ≥ 2.1%. However, in this study, mean FMD in both groups is ≥2.1% and change in FMD is only about 1.1% in the INHD group after 12 months follow up; although the change is statistically significant, its clinical relevance is still unknown. Our findings support the need for larger randomized trials with longer follow-up to determine whether INHD can decrease morbidity and mortality in patients who have ESRD and require maintenance HD by improving endothelial dysfunction.

The correlation of serum phosphorus and FMD seen in our study may provide a novel potential mechanism for beneficial effects of INHD on FMD. Compared with CHD (3 × 4 h), INHD (3 × 8 h) has a longer dialysis time and can provide enough dialysis doses to improve outcomes. Previous uncontrolled studies (17–19) of INHD have demonstrated decreased serum phosphorus, improved BP control and higher Hb levels. In our results, INHD had better control of serum phosphorus and BP than CHD. Recent data have demonstrated that dysregulation of phosphate homeostasis may contribute to the endothelial dysfunction. Hyperphosphatemia has been shown to induce acute endothelial dysfunction in vivo and exposure to a phosphorus load has been shown to increase reactive oxygen species production, induce apoptosis, and decrease nitric oxide (NO) production in endothelial cells in vitro (20,21). In a double-blind crossover study, FMD was measured before and 2 h after meals containing 400 mg or 1200 mg of phosphorus. The higher dietary phosphorus load increased serum phosphate at 2 h and significantly reduced FMD indicating a causal relation between endothelial dysfunction and acute postprandial hyperphosphatemia (22). Our study is the first to demonstrate that better control of phosphorus is associated with an improvement in FMD by INHD.

Some limitations deserve consideration. First, our study is a non-randomized control design, we cannot rule out the possibility of residual confounding, the baseline risk profile was more adverse among patients in INHD groups, and our findings may not be completely generalizable to all patients with ESRD or in other healthcare settings. The second is the techniques we have used to evaluate endothelial function. Changes in FMD have been described not only in the presence of classic risk factors, but also in a variety of other factors such as endogenous, exogenous, environmental, and familiar factors. This observation makes a correct serial evaluation of FMD in the same subject or a comparison of FMD between subjects very difficult because of the special attention that should be paid to all factors that can influence FMD (32). It is also operator-dependent. However, at present, this measurement is the preferred testing modality.

Antihypertensive treatment, especially ACE inhibitors and ARB, may influence FMD and influence the results in our study. Anderson et al. reported that ACEI could improve FMD in patients with coronary artery disease (33). Jawa et al. demonstrated beta-blocker treatment resulted in significant increase in FMD in African-American subjects with diabetes and hypertension (34). Cheetham et al. showed Losartan, an angiotensin type I receptor antagonist, improves conduit vessel endothelial function in Type II diabetes (35). Toma et al. showed that amlodipine may improve endothelial dysfunction in diabetes through anti-oxidant and anti-inflammatory mechanisms (36). In our study, the calcium antagonist taken by the subjects is amlodipine. At baseline, the proportion of patients who were taking antihypertensive drugs was similar and a strong influence of treatment on our results seems unlikely. At the end of follow-up, there was no significant change in antihypertensive medications in the CHD group, but the number of patients needed for antihypertensive therapy decreased significantly in the INHD group, and there were 15 patients who withdrew all antihypertensive medications. In the INHD group, with the reduced use of antihypertensive drugs, the protective effects on endothelial function of these drugs was less than that in the CHD group, but the endothelial function had still improved, which is helpful to explain that INHD is better than CHD in improving endothelial function.

CONCLUSION

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONCLUSION
  7. Acknowledgments:
  8. REFERENCES

In summary, compared with conventional hemodialysis, in-center nocturnal hemodialysis improves endothelial function; the better control of serum phosphorus is associated with the improvement of endothelial function.

Acknowledgments:

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONCLUSION
  7. Acknowledgments:
  8. REFERENCES

This work was supported by a grant from the Shanghai Municipal Health Bureau (2010062). The funder had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

REFERENCES

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONCLUSION
  7. Acknowledgments:
  8. REFERENCES