Aims: To prove the benefits of an specific strength training program in COPD patients compared to those achieved with the traditional endurance training. We will assess which is the proper training modality and which of them is the most beneficial according to the features of the patient to whom it is applied.

Subjects and Methods: We will study 66 patients diagnosed of COPD with a moderate-severe obstruction (FEV1<60%) and clinically symptomatic. The monitoring will be done for 3 moths. Patients will be differentiated between COPD with "predominant chronic bronchitis" and COPD with "predominant emphysema", with regard to clinical, functional and radiological criteria. After stratification of the sample, patients will be prospectively assigned to three groups of 22 patients each: a) Endurance training group, on cycle ergometer at a workload of 70% VO2max. b) Strength training group, with 5 different weightlifting exercises (4 sets of 6-8 repetitions). c) Mixed training group, with half the time dedicated to each type of training. Training will be developed during 12 weeks, three times per week, in 40 minutes sessions. Analysed variables, at the beginning and at the end of program will be: Chest X-ray, basic blood analyses, electrocardiogram, simple spirometry and bronchodilator test, blood gases, static pulmonary volumes, diffusion measurement and respiratory muscle pressure. Muscle strength will be assessed by means of 1RM test and the Myometer® dynamometer. Using the muscle skeletal ultrasound, we will control changes in muscle fibre size. Maximal and submaximal exercise tests on cycle ergometer will be also performed, so as a walking test (shuttle walking test). At last, we will assess treatment effect on dyspnea (Mahler's test) and quality of life (CRDQ).

Hypothesis: A strength training program would significantly increase peripheral muscle power. This type of training would mainly have effect on the "predominant emphysema" COPD patient, where the important weight lost causes a decrease in muscle strength.

1. Population: Patients with COPD diagnosed according to criteria established by the ATS (American Thoracic Society) and the SEPAR (Spanish Society of Pneumology and Thoracic Surgery) to submit a modereda-severe obstruction to airflow. (FEV1 <60%) and a clinical impact of their disease.

2. Sample: 66 patients diagnosed at the Department of Pulmonology, Hospital Virgen del Rocío de Sevilla, informed consent.

3. Exclusion criteria:

   • other cardiorespiratory diseases.
   • Systemic diseases
   • Inability or discomfort to participate in an exercise program

4. Study groups: patients will differ in COPD "prevalence chronic bronchitis " or COPD "prevalence emphysema, " according to the criteria allowed by the SEPAR and ATS.

   After stratification of the sample will be included in a prospective and randomized into three groups of 22 patients:

   Trained group 1, resistance 2 nd Group trained force 3 º Group Joint trained to strength and endurance

5. Variables: Parameters will be measured at baseline and end of follow-up period of 12 months.

   • CXR, elementary analysis, 12-lead electrocardiogram
   • simple spirometry and bronchodilator test
   • Blood gases
   • static lung volumes (FRC, RV, TLC) by plethysmography.
   • Study of diffusion (DLCO) by single breath technique
   • maximum muscle pressures both inspiratory and expiratory.

   CT scan (Phillips TomoScan with Easy Vision workstation version 4.2) which allows densitometric analysis of lung parenchyma.

   • Estimation of muscle strength:

     1. Repetition maximum test (Test 1 RM)
     2. Maximum isometric quadriceps strength and biceps measured with dynamometer (MIE Myometer ltd. with CAS software)
     3. Determination of muscular fatigue in the isometric tests of quadriceps and biceps, with the same system dynamometer (MIE Myometer ltd., software CAS)

   • Two-dimensional ultrasound of the quadriceps muscle and biceps to measure the size of muscle fibers. Probes will be used for external parts with a frequency of 7.5 MHz measurements were made from the skin depth to each of the different muscles and the thickness of each of them.

   • Stress Test:

     1. Maximum, with cycle ergometer and methodology already known (11), including: Determining, from the respiratory gases, oxygen consumption, carbon production and indirect anaerobic threshold, breathing pattern (minute ventilation, tidal volume, respiratory rate) , electrocardiography and pulse oximetry during exercise with heart rate control, blood pressure and dyspnea Borg scale.
     2. Submaximal cycle ergometer resistance with 80% of the load reached the ultimate test, with essentially control the resistance time, distance walked, and control of saturation (pulse oximetry) and heart rate achieved.
     3. Shuttle walking test, walk test is performed continuously walk between two points separated 10 meters and control the pace of a hearing (12).

   Assessment of dyspnea and quality of life:

   1. Baseline dyspnea index and index of transitional measures functional impairment, magnitude of the task and the magnitude of effort (Test of Mahler).
   2. Assessment of quality of life based on the specific questionnaire (CRQ) proposed by Guyatt for COPD patients and modified for your understanding in Spanish-speaking subjects.

6. Study design:

   The training of the different groups were carried out for 12 weeks, three times per week in sessions of 40 minutes duration.

   1. The resistance training will take place in cycle ergometer with a level of initial charge of about 70% of initial maximal oxygen consumption, increasing the load every two weeks as tolerated.

   2. The strength training with weight lifting practice and will consist of 4 sets of 6-8 repetitions of simple exercises. These are held at a station multigimnástica (CLASSIC Fitness Center, KETTLER) and are:

      • Simple bending of the arms (pull cable to the chest "Chest pulls"): Effects on the latissimus dorsi, deltoids and biceps. Sitting in the face of the tower of strength, approaching the bar to the chest and arm span later after return to the starting position.
      • Simple extension arms (force to the neck "Neck presses): Effects on the deltoids and triceps. Sitting upright on the bench and held the weight to shoulder height, arm span will be held over his head, slowly returning after bending arms to starting position.
      • arm flexion and extension against resistance (Mariposa "Butterfly"): Effects on the pectoral and deltoid muscles. Sitting upright on the bench with your forearms and elbows on a lever, takes it to the midline, giving way slowly to the starting position.
      • Extension legs (leg extension): Effects on quadriceps. Sitting on the bench leg extension is performed against resistance.
      • Leg Curl (leg curls): Effects on femoral biceps and calves. Lying prone on the bench legs are flexed against the peso.

      Resistance is increased gradually from 50% of the maximum weight you can lift the patient on one occasion, early in the week, up to 85%. Every two weeks the weight limit is reassessed to adjust the training load in each patient (Test 1 RM).

   3. In the group that carried out the joint training, half the time of each session will be devoted to strength training, following the same methodology (same number of years but with two sets of each), and the other half training resistance bike, beginning with one or the other interchangeably.

   Upon completion of 12 weeks of training will be repeated all baseline tests (except for chest CT) and compared the results.

7. Exit criteria of the study:

   • patient's clinical decompensation
   • Loss of 3 or 5 consecutive sessions staple
   • On the initiative of the patient.

8. Statistical analysis:

Analyzing the effect size at the expense of an improvement in work reached more than about 15 W in the stress test, considered a standard deviation of this parameter of ± 10 W, an alpha of 0.05 and a beta of 0.1 (statistical power 90%) and calculating a loss ratio of 15%, we estimate a sample size of 22 patients for each group.

Comparing before and after the training response in each group were performed using the paired t test. Comparisons between trained groups were made using analysis of variance (ANOVA). The Pearson correlation coefficient and linear regression analysis will allow to examine relationships between physiological variables. Be considered significant at p <0.05. The dispersion of a mean value is expressed as mean ± standard deviation.

Inclusion Criteria:

• Patients with COPD diagnosed according to criteria established by the A.. TS (American Thoracic Society) and the SEPAR (Spanish Society of Pneumology and Thoracic Surgery) to submit a modereda-severe obstruction to airflow. (FEV1 <60%) and a clinical impact of their disease.

Exclusion Criteria:

• other cardiorespiratory diseases.
• Systemic diseases
• Inability or discomfort to participate in an exercise program