Comparison of the distances covered during 3 and 6 min walking test

Vol. 96 (2002) 812^ 816 Comparison of the distances covered during 3 and 6 min walking test MILAGROS IRIBERRI,JUAN BTA GAŁ LDIZ,AMAIA GOROSTIZA,PEDRO ANSOLA AND CARMEN JACA Unit of Respiratory Pathology, Cruces Hospital, Baracaldo, Basque Country, Spain Abstract Aims: To determine the reproducibility of the distance covered in 3 min and its correlation with the 6 min walking test, as well as compare the distances covered at di erent time intervals. Secondly, to evaluate the relationship between the distances covered during these time periods and the maximum oxygen intake obtained during a bicycle ergometer test. Methods:Forty- ve Chronic Obstructive Pulmonary Disesase patients were included in the study. Subjects who were either physically limited or familiar with the test, or those with acute exacerbation in the month prior to the study, were excluded.three walking tests were carried out each day In 30 patients for three consecutive days, and the distances covered in periods of 3 and 6 min were measured with 20 min rest between each walk.no incentive was given and the patients knew thatthe distances coveredin 3 and 6 min would be quanti ed.oxygen saturation, heartrate and degree of breathlessness (modi ed Borg scale) were registered at baseline. After 3 min, the distance covered and degree of breathlessness were also measured. After 6 min, oxygen saturation, heart rate, degree of breathlessness and distance covered in meters were registered. Spirometry was performed daily on each patient, and those with an FEV 1 variation of less than10% were considered clinically and functionally stable. An exercise test using bicycle ergometer was carried outto determine maximum oxygen intake. A 3 min walking test was performed in15 patients, independently on the same day, whichwas followed after 20 min rest with a 6 min walking test. Results: A signi cantincrease was observed in the distance covered over 3 and 6 min in the rst 5 walks, with the greatest increase seen in the rst 3 walks.the correlation between the distance covered in 3 and 6 min was 0.98. The correlation between the distance covered in 3 min and oxygen intake was 0.64. No signi cant di erences were observed between the distances covered in the 0^3 and 3 to 6 min periods.during the walking test, breathlessness was measured using the modi ed Borg scale, which was 1.8 after 3 min, and 3.2 after six min and 8.6 atthe end of bicycle ergometer test.no signi cantdi erences were observed betweenthe distance covered during the 3 minute test and the distance inthe rst 3 min ofthe 6 min walking test. Conclusions: A learning e ect was observed when the walking test is carried out repeatedly over short time periods, with a signi cant increase in the rst 5 walks. Correlation between the distances covered in 3 and 6 min is very good, and acceptable when the distance covered over these periods is compared with oxygen intake and walking speed is constant.r 2002 Elsevier Science Ltd. Allrights reserved. doi:10.1053/rmed.2002.1363, available online at http://www.idealibrary.com on Keywords COPD; walking test. ABBREVIATIONS COPD = chronic obstructive pulmonary disease; VO 2 = oxygen intake, FEV 1 = forced expiratory volume in1sec. Received 9 March 2001, accepted in revised form15 March 2002. Correspondence should be addressed to: Milagros Iriberri Pascual, Laboratorio de Exploracio n Funcional, Unidad de Patolog a respiratoria, Hospital de Cruces, Plaza de Cruces s/n Baracaldo 48903, Vizcaya, Spain. Fax: 34 94 6006541; E-mail: jbgaldiz@hcru.osakidetza.net INTRODUCTION Since it was rst introduced in 1976 by McGavin (1), the walking test has been considered the standard test for the functional evaluation of patients with chronic obstructive pulmonary disease (COPD) and is used in diverse clinical situations; evaluation of pulmonary rehabilitation protocols, drug response or preoperative study of patients submitted to lung transplant or volume reduction. The test is simple, easy and non-incremental. Initially lasting 12 min, the test correlated well with max-

COMPARISON OF THE DISTANCES COVERED DURING 3 AND 6 MIN WALKING TEST 813 imum oxygen intake (VO 2 ) obtained in incremental tests using a cycle ergometer. More recently, the walking test has been shortened for greater comfort (2^5). Six minutes is presently considered standard for this type of test. It is known that, regardless of the duration, walking tests are in uenced by factors which modify the value of the results. Thus, the mood of patients (6), how the patient is motivated (7) and the learning e ect (7, 8) are factors that can some way modify the results.our objective was to analyze whether the distance covered over a short time period, 3 min, can be considered a reproducible test, determine whether there is a learning e ect, and correlate the results with those obtained from the standard 6 min test. Additionally, it was our aim to establish a correlation between the distance covered over 3 min and the maximum VO 2 obtained from bicycle ergometer. We feel that the possible use of the short walking test in the functional evaluation of COPD patients in an out-patient environment, as a complementary test to others such as forced spirometry, bronchodilation and arterial blood gases, fully justi es this study. PATIENTS AND METHODS A group of 45 clinically stable COPD patients (ATS criteria) (9), undergoing an out-patient-hospital rehabilitation program, were included in the study. Inclusion criteria included air ow obstruction with an FEV 1 /FVC o70% and FEV 1 o70%, with an FEV 1 reversibility lower than15% and less than 200 ml at15 min after inhalation of 200 mg of albuterol. Patients with acute lung exacerbation in the month prior to the study and those familiar with or physically limited to carry out the test were excluded. Study design We divided the patients into two groups. Group A, 30 patients. In this group a Spirometry (Sibel Medical, Barcelona,Spain) was performed for 3 consecutive days, using ERS standards (10). A variation of less than 10% in FEV 1 during the 3 consecutive days, with no change in normal medication throughout the period was all considered as the stability criteria. The walking test was performed in a 60 m long hospital corridor. Reproducibility was determined by carrying out the test for three consecutive days, three test walks per day with a 20-min rest between each walk. Each patient performed 9 walks and avoided visual or hearing encouragement of any kind whilst carrying out the test. Walking speed had to be the patient s maximum, who wasaskednottostop,buttosethisownwalkingpace. The patient was not accompanied by the physician during the walk. Patients were informed prior to the test that thedistancecoveredover3and6minwouldbequanti- ed.the following variables were recorded: At baseline: Heart rate and oxygen saturation (Omeda Biox 3740) was recorded, as well as degree of breathlessness using the modi ed Borg scale. At 3 min: Without stopping the patient from walking, the distance covered was recorded as well as the degree of breathlessness. At 6 min:the distance covered was recorded as well as heart rate, oxygen saturation and degree of breathlessness. A bicycle ergometer test (Oxycon-Delta Jaeger Wuerzburg, Germany), using a scaled incremental method, was performed no later than 1 week after the nal walking test, using steps of10 W/min. Limiting factors included symptoms, heart rate 90% greater than predicted, ratio of maximal ventilation (VE max ) to maximum voluntary ventilation (MVV) greater than 80% or decrease 44% in oxygen saturation. Data regarding maximal strength, maximum VO 2,ventilatory and hemodynamic parameters were obtained. Group B: In 15 patients, a 3 min walking test was performed and after 20 min rest a 6 min walking test was carried out.the distances covered both in the 3 min rst test as well as 0^3 and 0^6 min of the 6 min walking test were quanti ed. Statistical analysis The mean distances covered in subsequent tests were compared using variance analyses for repeated measurements. Thereafter, comparisons for paired tests (usingttests for paired data) were established in order to determine the minimum di erence of mean values with statistical signi cance. The correlation between variables was carried out using the Pearson test. Within-person standard deviation (7) was calculated for the distance variability over the period, and standard deviation between the mean was used to de ne the variance coe cient. A Po0.05 was considered signi cant. RESULTS Group A: A total of 30 patients, 25 male and 5 female with a meanage of63 + 7.9 years, were studied.the functional values showed moderate to severe air ow limitation, with a mean FEV 1 of12737311ml (43% of predicted value) and a FEV 1 /FVC of 39711. Other functional parameters were: TLC132731% predicted, FRC186742% predicted, TLCO 79723% predicted and an oxygen saturation of 9472atbaseline. The distance covered over 3 min increased throughout the di erent walks.in this regard, the mean distance covered in the rst walk was 225720 m, with a progressive increase until walk 9,251727 m, with an increase of 11% (26 meters). A signi cant increase in the distance

814 RESPIRATORY MEDICINE covered was observed in the rst 5 walks (Po0.001).The greatest increase was seen in primarily in the rst 3 walks, which increased 7.1%. The distance covered in 6 min was 453 + 41m in the rst walk, with a similar progressive increase in the following walks as was observed in the 3 min walk. Thus, the distance covered in walk 9 was 508757 m with an increase between the rst and ninth walk of 12% (55 m). A signi cant increase was observed in the distance covered over the rst 5 walks (Po0.001), and as in the 3 min test, the greatest increase (10%) occurred in the rst 3 walks. Mean distances were covered in the 6 min test throughout the 9 walks.the relationship between the distances covered at 3 and 6 min was 0.98 (Fig.1) (Po0.0001). The distance covered in accordance with 3 min time periods was evaluated from 0 to 3 min and 3 to 6 min. Walking speed was constant and no signi cant di erences were observed in any of the 9 walks between the distance covered in the rst period and that covered in the second 3^6min period as is shown in Fig. 2. Variability throughout the tests, expressed by the standard deviation in the same patient and variance coe cients, showed that at 3 min, this standard deviation was 11.41m, and 24.51 meters at 6 min. The variance coe cientforthedistancecoveredwas4.42%at3min,and 4.76% at 6 min (Fig. 3) The relationship between the distance covered at 3 min in the 5th walk and maximum oxygen intake was 0.64 (P = 0.001).This correlation was 0.61 (P =0.002)when compared with the distance covered at 6 min, with a maximum VO 2.The degree of breathlessness with modi ed Borg scale was 0.28 at baseline, 1.8 at 3 min, 3.2 at 6 min and 8.6 at the end of bicycle ergometer test. FIG. 2. Mean distance covered for the patients included in Group A, periods (0^3) and (3^6) min. No signi cant di erences were observed between the distances covered in both periodsin each walk. The correlation between the distance covered at 3 min in the 5th walk with pulmonary function parameters at rest was weak, 0.24 with FVC% and 0.21 with FEV%. Likewise, the distance covered at 6 min was also seen to slightly correlate with FVC (r = 0.27) and with FEV 1 (r =0.18). Group B: Fifteen patients were evaluated,10 males and 5females,meanage647 3yearswhoindependently performed a 3 and 6 min walking test. The distance covered in the 3 min test was 243728 m.in the rst 3 min of the 6 minute test, the distance covered was 241726 m, with no signi cant di erences (P = 0,164) between both distances. FIG. 1. Representation of the degree of correlation between the distance covered at 3 min and 6 min, R =0.98. DISCUSSION Our results suggest that the 3 min walking test is a submaximal strength test that is simple and easy to carry out both in the hospital as well as outpatients environment, with a low cost, facilitating the evaluation of the functional capacity of patients with COPD, not very severe patients. Potential use includes evaluation of results after therapeutic treatment: bronchodilator drugs therapy, lung volume surgery, lung transplant and respiratory rehabilitation programs. The results show that a learning e ect exists with the 3 and 6 min distances repeated over short periods of time. A good correlation was observed between the distance walked at 3 and 6min. The 6 and12 min walking tests are non-incremental efforts and tests of endurance, where the e ort carried by the patient depends on several factors which will inevitablya ectthedistancecovered,primarilyencouragement and the learning e ect (7). At present, no

COMPARISON OF THE DISTANCES COVERED DURING 3 AND 6 MIN WALKING TEST 815 FIG. 3. Comparison between the distance walked in 3 and 6 min, according to within-person standard deviation and coe cient of variation. consensus exists as to the use or not of such stimulation, since although a greater distance is observed, variability increases when this factor of not easy standardization is introduced. In our study, we decided to avoid the e ect of stimulation, making the test as simple as possible. Thus, all visual and audible stimulation on the part of personnel supervising the test in all walks were avoided and the patients were left to determine their own pace. In relation to learning and reproducibility e ects, consensus exists that an increase in the distance walked is observed when the test is repeated. Thus, McGavin (1) and Butland (2) showed that the learning e ect is seen primarily in the rst 2 walks. Knox (8) showed that this e ect was greater when the tests were repeated in shorter time periods.these di erences are smaller when walks are repeated over longer time periods, where increases are lower (8.5%). Our results with tests carried out in shorter time periods show a signi cant increase in thedistancecoveredover3and6min,thisincreasebeing more signi cant in the rst 5 walks. The increase in the distance walked between walks1and 9 at 6 min was 57 m (12%) and 26 m (11%) at 3 min. In our study, we compared whether the distances walked in both time periods, between 0 and 3 min and between 3 and 6 min were similar. As mentioned in methods, the test was designed so that the patient was aware beforehand that the distance walked at 3 and 6 min would be quanti ed in each period of time.our results show that the distances covered in each period were similar, with no signi cant di erences between each, in other words, walking speed was constant. There is a possibility that, in line with our study design, the patients may have chosen a constant walking pace during the 6 min test, since they were already aware that thedistancecoveredwouldbemeasuredat3and6min. This could, in part, justify our results which show a similarity in the distance covered in both the 0 ^3 and 3^ 6 min periods. To answer this question, we compared the distance covered by a group of 15 patients in a 3 min walking test with the rst 3 min of a 6 min walking test and we observed that both distances were practically the same, 243 7 28 vs 241 726 with no signi cant di erences.this would suggest that both distances walked are the same independently if the subject knew previously the duration of the test. Bernstein et al. (5) also have likewise evaluated the distance between di erent periods (2 min), with patients performing a 12 min walking test. In that study, the patient was simply asked to walk the furthest distance possible in that period of time (12 min). They found no di erences in the distance covered at di erent 2-min time periods, suggesting that patients maintain the same pace throughout the 12 min test, without signi cant differences in the di erent periods. The authors, by way of criticism, stated that had the patients known beforehand the duration of the test, the results at di erent time periods would have been di erent. Due to the design of our study, this problem did not exist. Our study analyzed the within person variability (7) of the walking test in each patient, quantifying standard deviation throughout 9 walks.with the 3 min walk, this was 11.41, and 24.51 with the 6 min walk, suggesting good reproducibility with no real dispersion of the distances by each patients walks. These data are similar to that observed by other authors (7) who found values of 7.3 m for the distance walked in the 3 min test, and 22.25 for the distance covered in 6 min. In our study, the variance coe cient was 4.42% at 3 min and 4.76% at 6 min, suggesting good reproducibility in the distance walked for both time periods. One important point for consideration is the clinical signi cance of the di erence in distances walked by the

816 RESPIRATORY MEDICINE same patient, which is fundamental in protocols such as respiratory rehabilitation. Redelmeier (11), who studied 112 stable COPD patients with the aim of determining the di erence in distance during 6 min walking test capable of discriminating changes in functional capacity with clinical signi cance, felt that this distance was 54 meters. At present, there are no studies that have used a walking test lower than 6 min to evaluate the variation in the distance covered with clinical signi cance, suggesting a possible limitation of such tests. However, authors such as Strijbos (12) have used a 4 min walking test in home and hospital pulmonary rehabilitation programs, and have found the test to be sensitive for evaluating di erences from a statistical point of view. Although the 6 min walking test is, at present, considered a low-cost, objective, safe and simple, it is generally agreed that the direct method for determining a maximal e ort is to carry out a bicycle ergometer test and determine maximum oxygen intake. The correlation between the distance walked at 3 and 6 min with maximum VO 2 was lower and similar in both times periods, 0.64 at 3 min and 0.61 at 6 min. These values are in the same range other authors have recorded (13, 14). Bernstein (5) observed that the degree correlation between maximum VO 2 and the distance covered in the walking test is lower when the latter is shortened (r =0.53at2min,r =0.59at4min,r =0.64at6min, and r = 0.72 at 12 min). Walking tests can be considered submaximal and more physiological tests of endurance, since they represent the patient s daily activities best. At present, and due to the di erences between VO 2 data obtained from walking tests in relation to strength tests, as well as the di culties in reproducibility of the former, several groups (15,16) suggest carrying out the walking test with progressive loads (Shuttle Walking Test), since theoretically it is more reproducible and correlates better with maximumvo 2.The Shuttle test is an incremental test, not an endurance test. The walking speed is not determined by the patient but rather it is imposed with increasing paces. These characteristics make this walking test conceptually more like a maximal incremental test, carried out using bicycle ergometer and which explain the better relationship obtained in variables such as VO 2 (17,18,19), but in our opinion is clearly more complex requiring greater resources and more time. The use of short test in out-patient rehabilitation programs in COPD patients has already been carried out by groups, such as Strijbos (12), who showed that simplifying the walking test is possible in out-patient studies and can be performed with few resources and in a short periods of time. Presently, an ever increasing number of authors (20, 21) include the walking test as routine test along with spirometry, bronchodilation, arterial blood gases for the evaluation and classi cation of COPD patients.this must favor the use of simple tests that are easy to perform, especially when bearing in mind the potential number of patients requiring evaluation. REFERENCES 1. McGavin CR,Gupta SP, McHardy GJR.Twelve-minutewalking test for assessing disability in chronic bronchitis. Br Med J1976; 1: 822^823. 2. Butland RJA, Pang J,Gross ER et al.two,six,and12minutewalking test in respiratory disease. Br Med J1982; 284:1607^1608. 3. Swinburn CR, Wake eld JM, Jones PW, et al. Performance, ventilation, and oxygen consumption in three di erent types of exercise test in patients with chronic obstructive lung disease.thorax 1985; 40:581^586. 4. Edgar PAW, Van Heurn LWE, Morgan AD, et al. Choice of walking test asses respiratory disability and breathlessness.thorax1984; 39: 226. 5. Bernstein ML, Despars JA, Shing NP, et al.reanalysisofthe12minute walk in patients with chronic obstructive pulmonary disease. Chest 1994; 105: 163 ^167. 6. Morgan AD, Peck DF, Buchanan DR, et al. E ect of attitudes and beliefs on exercice tolerance in chronic bronchitis. Br Med J 1983; 286:171^173. 7. Guyatt GH, Pugsley SO, Sullivan MJ, et al. E ect of encouragement on walking test performance.thorax 1984; 39: 818^822. 8. Knox AJ, Morrison JFJ, Muers MF, et al. Reproducibility of walking test results in chronic obstructive airway disease.thorax 1988; 43: 388^392. 9. Celli,B,Snider G,He ner J,et al. ATS Statement. Standards for the diagnosis and care of patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med 1995; 152: s77^ s129. 10. Report working party standardization of lung function tests European community for steel and coal. Lung volumes and forced ventilatory ows. EurRespirJ1993; 6: 5^40. 11. Redelmeier DA, Bayoumi AM,Goldstein RS, et al. Interpreting small di erences in functional status: the six-minute walk test in chronic lung disease patients. Am JRespir Crit Care Med1997:155:1278^1282. 12. Strijbos JH, Postma DS, Altena RV, et al. A comparison between an outpatient hospital based pulmonary rehabilitation program and a home-care pulmonary rehabilitation program in patients with COPD.Chest 1996; 109: 366^372. 13. Nakagaichi M, Tanaka K. Development of a 12 min treadmill walk test at a self selected pace for the evaluation of cardiorespiratory tness in adult men. Appl Human Sci 1998; 17: 281^288. 14. Cahalin LP, Semigran MJ, Dec GW. Assessment of oxygen uptake during the six minute walk test.chest 1997; 111: 1465^1466. 15. Elias MT, Ferna ndez J, Toran J, et al. Reproducibilidad de un test de paseo de carga progresiva (shuttle walking test) en pacientes con enfermedad pulmonar obstructiva cro nica. Arch Bronconeumol 1997; 33: 64^68. 16. Shing SJ, Morgan MD, Scott, et al. Development of a shuttle walking test of disability in patients with chronic airways obstruction.thorax 1992; 47:1019 ^1024. 17. Elias MT,Ortega F, Ferna ndez J, et al.comparacio n de un test de carga progresiva (shuttle walking test ) con una prueba de esfuerzo en cicloergo metro en pacientes con EPOC. Arch Bronconeumol 1997; 33: 498^502. 18. Shing SJ, Morgan MDL, Hardman AE, et al. Comparison of oxygen uptake during a conventional treadmill test and the shuttle walking test in chronic air ow limitation. Eur Respir J1994; 7: 2016^2020. 19. Revill SM, Morgan MDL, Shing SJ, et al.the endurance shuttle walk: a new eld test for the assessment of endurance capacity in chronic obstructive pulmonary disease.thorax 1999; 54: 213^222. 20. Eiser NM, Willsher D, et al. Comparison of walking test in COPD. EurRespirJ1999; 14 (Suppl 30): 422s. 21. Celli BR. Standards for the optimal management of COPD. A summary.chest 1998; 113: 283s^287s.