versión impresa ISSN 0102-3586
J. Pneumologia v.29 n.6 São Paulo nov./dic. 2003
Audrey. Borghi SilvaI; Valéria Amorim Pires di LorenzoII; Maurício JamamiI, Luciana. Maria. Malosá SampaioI; Aureluce DemonteIII; Leonardo CardelloIII, Dirceu CostaI
and Spirometry Laboratory of the UFSCar, SP
IICardiorespiratory Unit of the UNIARA
IIIFood and Nutrition Department UNESP Araraquara, SP
Patients with chronic obstructive pulmonary disease (COPD) usually present
intolerance to physical exercise due to ventilatory limitation. L-carnitine
has been used to enhance aerobic capacity in patients with chronic diseases,
but no studies have been conducted involving COPD patients.
OBJECTIVE: To evaluate the influence of L-carnitine supplementation (2 g/day) on COPD patients undergoing physical training (PT) 3 times a week for 6 weeks.
METHODS: Patients (mean age 69 ± 7 years, n = 30) with stable COPD and < 65% of predicted forced expiratory volume in 1 second (FEV1) were separated into 3 groups of 10 patients each. Group 1 (G1, n = 10) received PT and L-carnitine (2 g/day), Group 2 (G2, n = 10) received PT and placebo, and Group 3 (G3, n = 10) received only L-carnitine (2 g/day). Spirometry and a 6-minute walking test (6MWT) were performed before and after intervention. Plasma levels of free carnitine were measured at the beginning and end of the study.
RESULTS: A significant increase in walking distance was found in G1 and G2 patients (421 ± 100 to 508 ± 80.7 and 496 ± 78.7 to 526 ± 64.3 meters). However, the mean improvement on 6MWT was greater in G1 than in G2 (p < 0.05). In addition, heart rates during PT sessions were found to be significantly lower in G1. There were no significant changes in spirometric variables, oxygen saturation or dyspnea in any group. Plasma levels of free L-carnitine were found to increase only in G3 (59.2 ± 13.8 to 102.3 ± 15.3 mmol/L).
CONCLUSION: Oral L-carnitine supplementation combined with resistance training may improve tolerance to physical exercise in COPD patients.
Key words: Chronic obstructive pulmonary disease. Carnitine, therapeutic use. Spirometry, methods. Exercise. Walking.
Acronyms and abbreviations
used in this paper
ATS American Thoracic Society
CV- Vital capacity
FVC Forced vital capacity
SVC Slow vital capacity
COPD Chronic obstructive pulmonary disease
HR- Heart Rate
BMI Body Mass Index
PR Pulmonary rehabilitation
SO2 Oxygen saturation
WT6 6 minutes walk test
PT Physical training
FEV1 Forced expiratory volume at 1st second
MVV Maximum voluntary ventilation
Intolerance to physical exercise is a consequence of damage caused by chronic obstructive pulmonary disease (COPD) and occurs as a function of decreased ventilatory capacity combined with increased demand. One of the objectives of physical training programs used in pulmonary rehabilitation (PR) is to improve tolerance to physical exercise, reducing the symptoms of dyspnea, as well as increasing walk-test distance and tolerance to greater workloads.(1-4)
Decreased ventilatory capacity causes adaptive changes to the skeletal muscle system, reducing physical capacity. Skeletal muscle dysfunction is a factor that contributes to intolerance to physical exercise. The process involves adaptive changes such as decreases in capillarity and in the number of oxidative enzymes, thereby reducing aerobic capacity. In addition, the physical limitation resulting from COPD may be related to right ventricular dysfunction.(5) In a hemodynamic monitoring study, Mahler et al.(6) showed that, during physical exercise, pulmonary hypertension is a restricting factor that leads to interruption of exercise in patients with advanced forms of the disease. The right ventricular ejection fraction in patients with COPD may be reduced during exercise.(7) This may be due to either pulmonary hypertension or dynamic hyperinflation. As a result, blood flow to the exercised muscles would decrease, accentuating lactic acidosis and leading to interruption of the exercise, thereby deterring the rehabilitation process. The vitamin L-carnitine is a nutritional supplement that has been suggested to improve aerobic performance in patients with chronic disease.(8-10) This compound is a natural component of the organism acquired through animal protein intake and may be synthesized by the liver, kidneys and brain. Levels of L-carnitine in tissue are influenced by factors such as malnutrition, age, gender and some pathologies.(11) In addition, L-carnitine has been used as a supplement with proven positive results in enhancing aerobic performance.(12) This is probably due to its ability to facilitate the entrance of fatty acids into the mitochondria, thereby serving as as energy substrate for beta-oxidation.(11) Some studies show that L-carnitine could diminish muscle fatigue and pain caused by maximal physical exertion.(13)
Recent studies, in which L-carnitine was continuously given to patients with congestive heart failure, corroborated a significant improvement in tolerance to physical exercise and in patient physiology.(14,15) In patients with peripheral arterial disease, L-carnitine increased tolerance to physical exercise during the 6-minute walk test and improved muscle strength in the lower limbs.(9,16) Notwithstanding the encouraging results achieved in various clinical conditions, the authors are not aware of any previous assessment of the effects of L-carnitine on the exertion capacity of patients with COPD. This study intends to assess the effect of L-carnitine supplementation, combined with supervised physical training, on the tolerance to exercise of patients with stable COPD.
In this study, we included patients who had been clinically diagnosed with COPD, who presented a forced expiratory volume in the first second (FEV1) <65% of the predicted value, and who were clinically stable, over 50 years old and sedentary. We evaluated 40 COPD patients of both genders, and 30 were included. Ten patients were excluded due to respiratory instability, severe heart disease, or orthopedic problems that hindered physical training. In compliance with Resolution 196/96 of the National Health Council (NHC), patients who gave formal written consent to the objectives of the study were submitted to experimental treatment. This study was approved by the Ethics Committee on Human Research of our institution.
The sample was randomly separated into 3 groups. Patients in group 1 (G1, n= 10) were given supplementation of 2g/day of L-carnitine, ingested in 2 daily doses, and subjected to physical training (PT). Patients in group 2 (G2, n=10) were given a placebo and also subjected to PT. Patients in group 3 (G3, n=10) were given L-carnitine supplement for 6 weeks but did not perform PT. The L-carnitine was given in vials containing 1g. The placebo was similar in color and taste and was given in vials identical to those containing the L-carnitine. Prior to use, concentration of 1 of the vials was assessed through enzymology and found to correspond to that of the label. Although patients did not know if the ingested substance was the placebo or the L-carnitine, the researcher in charge of carrying out functional measurements was privy to this information. All supplementation, placebo administration and PT continued for a 6-week period.
Spirometry: After height and weight were measured, spirometry was performed using a Vitalograph model Hand Held 2021 spirometer (Ennis, Ireland). During tests in an air conditioned room, patients remained seated and carried out the maneuvers of slow vital capacity (SVC), forced vital capacity (FVC) and maximum voluntary ventilation (MVV). Technical procedures were carried out and criteria of acceptability and reproducibility were set according to the standards recommended by the American Thoracic Society.(17) Reference values used were those suggested by Knudson et al.(18) Three forced expiratory curves were produced, technically acceptable for measurement of FVC and FEV1, and MVV was directly obtained during the 3 maximal expirations during 15s. Results achieved were expressed in terms of temperature and pressure. Ventilatory disorders were classified as mild, moderate or severe, as designated by the Brazilian Consensus on Spirometry.(19)
Walk test: A 6-minute walk test (6MWT) was carried out in a level corridor 28 meters in length. A portable pulse oximeter (8500-A; Nonin Medical, Plymouth, MN, USA) was used to monitor patients during the entire test. Patients were questioned about dyspnea at the onset and at the end of the test using the Borg Scale to rate perceived exertion at Category Ratio 10, at which zero is considered as no shortness of breath and 10 as a very, very intense shortness of breath.(20) In an attempt to avoid allowing learning in the test to interfere with the results and in an endeavor to assure the greatest reliability of the results the 6MWT was performed twice for each patient during pretreatment. These 6MWTs were performed on alternate days, with the highest value used for statistical analysis of the data. The examiner oriented and stimulated patients at the beginning and during the test to walk as fast as possible, and standard verbal encouragement was given.(21) Values of hemoglobin oxygen saturation (SaO2) were monitored during the 6MWT, and the Borg Scale was used at the beginning and end of the test. For data analysis among the 3 groups, the variation (Delta) between the final and the initial distance (post-pre/pre) of the distance walked x 100 was utilized.
Analysis of plasma levels of free L-carnitine: Five 5-ml blood samples were collected at the beginning and end of the experiment in order to quantify free L-carnitine in the plasma. Blood was immediately centrifuged at 2500 rpm and stored in freezer at 80 C. Samples were stocked and analyzed in duplicate. For serum quantification of L-carnitine, enzymology(22) involving an Ultrospec C100 UV spectrophotometer (Pharmacia, Cambridge, England) was used.
Physical training: Patients from G1 and G2 underwent a program of physical training (PT) comprised of 10-min torso stretching sessions, upper and lower limb exercises and 30-min walking exercises on a treadmill. The training speed was 80% of the maximum achieved on a treadmill exertion test, with incremental increases of 0.5 km/h every 2 minutes and a constant 3% inclination, continuing until exhaustion. In addition, active, free exercises of upper and lower limbs with the aid of a baton were performed in standing, sitting or supine positions and were followed by relaxation exercises. This training program lasted for 1 hour and was conducted on alternate days, 3 times a week, for 6 consecutive weeks. When necessary, PT sessions were preceded by inhalation therapy and bronchial hygiene therapy. Patients in G3 (control group) were examined weekly and received only inhalation therapy or obstruction clearance according to the requirements of each patient. Cardiac rate (CR) was monitored continuously during each PT session and the exercise load was increased weekly.
Statistical analysis: For statistical analysis of data, the paired Student t-test was used to compare data obtained from each group during pretreatment and posttreatment. For analysis of the mean CR of the variation (Delta) in CR and velocity between the 2 PT groups, Mann-Whitney analysis was used. Since data presented normal distribution and equivalent variances, ANOVA was used to detect differences between the groups. When a significant difference was detected, Duncans post-hoc test was used to arrange the variables. The variation (Delta) between the final and initial values (post-pre/pre) x 100 was also assessed using ANOVA. The adopted statistical probability was of a p < 0.05.
The anthropometric and demographic characteristics of the 3 groups are described in Table 1. No differences were found among groups. At the initial spirometry evaluation, which was similar among the groups, airflow obstruction was mild in 2 patients, moderate in 4 and severe in 4. No statistically significant differences were found between pre- and posttreatment in terms of FVC, FEV1 and MVV (Table 2).
In the 3 groups of patients evaluated, no negative effects of L-carnitine or placebo were detected. In the 6MWT, no significant difference was found among the groups between distance walked in the test baseline values. However significant increases were found in the distance walked by the patients in G1 (421 m ± 100 m vs. 508 m ± 80.7 m) and by those in G2 (496 m ± 78.7 m vs. 526 m ± 64.3 m) (p<0.05). Patients of G3 did not present significant increases in the distance walked at re-evaluation (Table 2). In G1, 80% of the patients had increases greater than those considered clinically significant (54 m).(23) This increase was found in 40% of G2 patients and 30% of G3 patients. Figure 1 illustrates the variation (Delta) between the pre- and posttreatment distances walked in the 6MWT by all 3 groups. A greater increase in posttreatment distance walked was seen in G1 (p<0.05), where, by arrangement of the data (Duncans test): G1<G2 = G3.
With regard to CR in the 2 PT groups, the mean CR achieved during the 6 weeks was not statistically different between G1 and G2. However the variation (Delta) between the initial and final CR was different between the 2 groups, showing significantly decreased CR in G1 only, although weekly velocity increased at a similar rate in both groups (Figure 2). No significant alteration in the Borg Scale SaO2 scores was detected during the 6MWT for any of the groups under study, nor in the comparison of these same variables among the groups.
With regard to the variation in serum fractions of L-carnitine in the patients studied, only patients in G3 showed a statistically significant increase (p<0.05).
The current study evaluated the effects of a supervised program of physical training, with or without L-carnitine supplementation, on the tolerance to exercise (6MWT) of patients with COPD. It was noted that supplementation with L-carnitine combined with physical training improves performance on the 6MWT and lessens CR. Furthermore, higher levels of plasma L-carnitine were found in patients who received L-carnitine supplementation of and did not perform PT than in those receiving L-carnitine and PT together.
Among the various tests adopted for evaluating of functional status during and after physical training programs, the 6MWT has, in several studies, proven to be easy to perform, reproducible and useful for the evaluation of tolerance to physical exercise in patients with COPD(23) after a supervised PT program.(24,25) According to our results, there were significant increases in distance walked among patients in the PT groups (Table 2). However patients that received PT and L-carnitine supplementation presented a greater gain in distance walked than did patients in the other 2 groups (Figure 1).
It has been proven that even short PT programs can increase tolerance to exercise in COPD.(1,26) Nevertheless, few studies combining the effects of supplementation of compounds with physical training have reported positive performance results. For instance, Sridhar et al.(27) gave supplementation to a group of patients with COPD in a pulmonary rehabilitation program with increasing nutritional intake over a 4-month period but did not achieve significant improvement in anthropometric measurements, pulmonary function, respiratory muscle strength or maximal oxygen consumption. Administration of anabolic steroids(28) to undernourished patients (aiming to increase lean muscle mass) has also been studied, although capacity to exercise did not change. In another study, supplementation with branched-chain amino acids for a period of 3 months did not significantly increase body weight and respiratory muscle strength.(29)
Nutritional supplementation has been thoroughly investigated in COPD patients, mainly in depleted ones. However, according to Ferreira et al.,(30) doubts still persist concerning the efficacy of supplementation in improving anthropometric measurements and in the functional capacity of stable COPD patients to exercise. Schols and Wouters,(31) propose that additional studies be carried out to investigate whether a metabolic response is enhanced by modulation of muscle cell metabolism by so-called "bioactive nutrients" such as creatine, carnitine, the anti-oxidants and amino acids.
Although velocity increase might have been similar in both PT groups (G1 and G2) during this study, it was noted that, during the PT sessions on the treadmill, CR only decreased in G1. This suggests that G1 patients may have suffered less cardiovascular overload when test time and work velocity increased. As previously discussed, L-carnitine supplementation combined with PT seems to have had a beneficial effect on the aerobic performance of the G1 patients. Another possible hypothesis is that the muscle oxidative metabolism may been enhanced, since lower CR may be related to better peripheral oxygen extraction.
Previous studies have reported that nutritional supplementation alone can have positive effects on the cardiovascular system in patients with chronic disease. Some authors(32,33) have shown that isolated L-carnitine supplementation increased physical performance in patients with cardiovascular diseases. In our study, it was noted that, when the 3 groups were compared for performance at 6MWT, G2 and G3 were similar, suggesting that isolated supplementation may result in enhanced performance in COPD patients. Despite the fact that there was a certain agreement with data in literature, such results must be cautiously considered, since the gains achieved in G2 and G3 were very small when compared to those of G1.
In the current study, increases in plasma levels of free L-carnitine were found only for G3 (isolated supplementation), whereas those values were unchanged in G1. This finding is in agreement with data from Heinonen,(34) who previously observed the same result and speculated that the PT resulted in greater L-carnitine activity as a carrier of fatty acids to serve as an energy substrate. Some authors(9,32) have reported that, in habitual exercise, the level of carnitine in muscle increased while plasma values decreased.(32) It is probable that these levels did not rise in the G1 patients due to the imposed chronic aerobic treatment, which permitted improved utilization of the L-carnitine as a carrier, thereby improving aerobic metabolism. Another probable explanation for the results is that free L-carnitine had been assimilated to the skeletal or cardiac muscle, resulting in a better final performance, although, in our study, muscle L-carnitine was not quantified.
We did not detect significant changes in FEV1, FVC or MVV in any of the patients under study. These findings were duplicated in literature, but only in patients submitted to PT alone.(22) Nevertheless, Vogiatzis et al.(24) compared COPD patients with mild, moderate and severe obstruction submitted to a PT program for 12 weeks and found significant increases in FEV1 and FVC. According to these authors, the increases in spirometry values represented a clinical improvement, as well as a more appropriate use of the inhalation therapy.
In this study, no changes in SaO2 and dyspnea were observed after PT. Some authors(2,35) have noted that there were no changes in SaO2 when COPD patients performed only high-intensity exercises. After the 12-week PT program used in our study, Dyspnea, as assessed using the Borg scale.(16), did not improve, although other studies have demonstrated lesser degrees of dyspnea after a PT program. (24)
The present study presented some limitations, such as the single-blind design of the L-carnitine and placebo administration, the small sample size and the absence of more complete physiological data. This implies the need for further studies. Nevertheless, our preliminary results show that this nutritional strategy may play an adjuvant role in the pulmonary rehabilitation of patients with COPD.
We wish to thank Prof. Regina Vendramini and Prof. Dr. Heloisa Sobreiro Lelistre for the orientation regarding collection and storage of the biological material. We would also like to thank Prof. Dr. Jorge Oishi for his assistance in the statistical analysis. In addition, we are grateful to Sintofarma Laboratories for supplying a portion of the L-carnitine used and to the Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) for providing financial support (grant no. 00/00311-6).
1. Boueri FMV, Bucher-Bartelson B, Glenn K, Make BJ. Quality of life with a generic instrument (Short Form-36) improves following pulmonary rehabilitation in patients with COPD. Chest 2001;119:77-84. [ Links ]
2. Casaburi R, Porszasz J, Burns MR, Carithers ER, Chang RS, Cooper CB. Physiologic benefits of exercise training in rehabilitation of patients with severe COPD. Am J Respir Crit Care Med 1997;155:1541-51. [ Links ]
3. Neder JA, Nery LE, Cendon Filha SP, Ferreira IM, Jardim JR. Reabilitação pulmonar: fatores relacionados ao ganho aeróbio de pacientes com DPOC. J Pneumol 1997;23:115-23. [ Links ]
4. Casaburi R. Skeletal muscle dysfunction in chronic obstructive pulmonary disease. Med Sci Sports Exerc 2001;33:662-70. [ Links ]
5. Olopade CO, Beck KC, Viggiano RW. Exercise limitation and pulmonary rehabilitation in COPD. Mayo Clin Proc1992;6:144-57. [ Links ]
6. Mahler DA, Brent BR, Loke BL, Zaret BL, Matthay RA. Right ventricular performance and central hemodynamics during upright exercise in patients with chronic obstructive pulmonary disease. Am Rev Respir Dis1984;130:722-9. [ Links ]
7. Sietsema K. Cardiovascular limitations in chronic pulmonary disease. Med Sci Sports Exer 2001;33(Suppl):656-61. [ Links ]
8. Anand I, Chandrashekhan Y, Giuli FD, Pasini E, Mazzoletti A, Confortini R, et al. Acute and chronic effects of propionyl-L-carnitine on the hemodynamics, exercise capacity, and hormones in patients with congestive heart failure. Cardiovasc Drugs Ther 1998;12:291-9. [ Links ]
9. Barker GA. Effects of propionyl-L-carnitine on exercise performance in peripheral arterial disease. Med Sci Sports Exerc 2001;33:1415-22. [ Links ]
10. Brass EP, Scharon MD, Sietsema KE, Hiatt WR, Orlando AM, Amato A. Intravenous L-carnitine increases plasma carnitine, reduces fatigue and may preserve exercise capacity in hemodialysis patients. Am J Kidney Dis 2001;37:1018-28. [ Links ]
11. Leibovitz B, Mueller J. Carnitine. J Optimal Nutr 1993;2:90-109. [ Links ]
12. Brass EP. Supplemental carnitine and exercise. Am J Clin Nutr 2000; 72(Suppl):618-23S. [ Links ]
13. Reynier M, Brun JF, Orsetti A. Metabolic effects of L-carnitine, L-lysine methionine during maximal exercise. Sci Sport 1996;13:17-25. [ Links ]
14. Loster H. Prolonged oral L-carnitina substitution increases bicycle ergometer performance in patients with severe, ischemically induced cardiac insufficiency. Cardiovasc Drugs Ther 1999;13:537-46. [ Links ]
15. Iyer RN, Khan AA, Gupta A, Vajifdar BU, Lokhandwala YY. L-carnitine moderately improves the exercise tolerance in chronic stable angina. J Assoc Phys India 2000;48:1050-2. [ Links ]
16. Watanabe S, Ajisaka R, Masuoka T, Yamanouchi T, SaitouT, Toyama M, et al. Effects of L and DL-carnitina on patients with impaired exercise tolerance. Jpn Heart J 1995;36:319-31. [ Links ]
17. American Thoracic Society. Pulmonary rehabilitation. Am J Respir Crit Care Med 1999;159:1666-82. [ Links ]
18. Knudson RJ, Leibowitz MD, Holberg CJ, Burrows B. Changes in the normal maximal expiratory flow-volume curve with growth and aging. Am Rev Respir Dis 1983;127:725-34. [ Links ]
19. Sociedade Brasileira de Pneumologia e Tisiologia. I Consenso Brasileiro sobre Espirometria. J Pneumol 1996;22;105-64. [ Links ]
20. Borg GA. Psychophysical bases of perceived exertion. Med Sci Sports Exerc 1982;14:377-81. [ Links ]
21. American Thoracic Society Statement. Guidelines for the six-minute walk test. Am J Respir Crit Care Med 2002;166:111-7. [ Links ]
22. Pearson DJ, Chase JFA, Tubbs PK. Methods enzymology 1968;14: 612-25. [ Links ]
23. Ribeiro SA, Jardim JRB, Nery LE. Avaliação da tolerância ao exercício em pacientes com DPOC. Teste de caminhada de seis minutos versus cicloergometria. J Pneumol1994;20:112-6. [ Links ]
24. Vogiatzis I, Williamsom AF, Miles J, Taylor IK. Physiological response to moderate exercise workloads in pulmonary rehabilitation program in patients with varying degrees of airflow obstruction. Chest 1999; 116:5:1200-7. [ Links ]
25. Brunetto AF, Paulin E. Melhora da performance física após fisioterapia respiratória em pacientes com DPOC. Rev Bras Fisioter 1998;3: 29-33. [ Links ]
26. Miyahara N, Ryosuke E, Takeyama H, Kunichika N, Moriyama M, Aoe K, et al. Effects of short-term pulmonary rehabilitation on exercise capacity and quality of life in COPD.Acta Med Okayama 2000; 54:179-84. [ Links ]
27. Sridhar MK, Galloway A, Lean ME, Banham SW. An out-patient nutritional supplementation programme in COPD patients. Eur Respir J 1994;7:720-4. [ Links ]
28. Ferreira IM, Verrescchi IT, Nery LE, Goldstein RS, Zamel N, Brooks D, et al. The influence of 6 months of oral anabolic steroids on body mass and respiratory muscles in undernourished patients with chronic obstructive pulmonary disease. Chest 1998;114:19-28. [ Links ]
29. Yoneda T, Yoshikawa M, Fu A, Tokuyama T, Okamoto Y, Fukuoka K, et al. Clinical benefit of nutritional assessment and support in patients with COPD. Nihon Kyobu Shikkan Gakkai Zasshi 1996;34:79-85S. [ Links ]
30. Ferreira IM, Brooks D, Lacasse Y, Goldstein RS. Nutritional support for individuals with COPD A meta-analysis. Chest 2000;117:672-8. [ Links ]
31. Schols AM, Wouters EF. Nutritional abnormalities and supplementation in chronic obstructive pulmonary disease. Clin Chest Med 2000;21: 753-62. [ Links ]
32. Brevetti G, Chiariello G, Ferulano A, Policicchio A, Nevola E, Rossini A, et al. Increases in walking distance in patients with peripheral vascular disease treated with L-carnitine: a double-blind, cross-over study. Circulation 1988;77:767-73. [ Links ]
33. Hiatt WR, Regeinsteiner JG, Creagger MA. Propionyl-L-carnitine improves exercise performance and functional status in patients with claudication. Am J Med 2001;110:616-22. [ Links ]
34. Heinonen OJ. Carnitine and physical exercise. Sport Med 1996;22: 109-32. [ Links ]
35. Lennon DLF, Shrago E, Madden M, Nagle F, Hanson P, Zimmerman S. Carnitine status, plasma lipid profiles, and exercise capacity of dialysis patients: effects of a submaximal exercise program. Metabolism 1986;35:728-35. [ Links ]
36. Berry MJ, Adair NE, Sevebsky KS. Inspiratory muscle training and whole body reconditioning in chronic obstructive pulmonary disease. Am J Respir Crit Care Med 1996;153:1812-6. [ Links ]
Address for Correspondence
Rod. Washington Luís, km 235
13565-905 São Carlos, SP Brasil
Tel.: (16) 260-8343
Submitted for publication on Dec/16/02. Accepted after review on Sept./11/03.
* Work carried out at the Universidade Federal de São Carlos, São Paulo, SP . Support: FAPESP (São Paulo State Foundation for the Support to Research) Proc. 00/00311-6