Gustavo J. Rodrigo, MD; Mario Rodriquez Verde, MD; Virginia Peregalli, MD; and Carlos Rodrigo, MD

Effects of Short-term 28% and 100% Oxygen on PaCO 2 and Peak Expiratory Flow Rate in Acute Asthma* A Randomized Trial Gustavo J. Rodrigo, MD; Mario Rodriquez Verde, MD; Virginia Peregalli, MD; and Carlos Rodrigo, MD Study objective: We conducted the first randomized controlled study to assess the effects of short-term 28% and 100% oxygen on PaCO 2 and peak expiratory flow rate (PEFR) in patients with acute severe asthma. Patients and interventions: Seventy-four patients (mean age, 37.9 9.7 years [ SD]; PEFR, 41.0 12.1% of predicted) from two emergency departments were randomized to receive 28% or 100% oxygen during 20 min. Results: The administration of 100% oxygen significantly increases PaCO 2 (p 0.03) and decreases PEFR (p 0.001) as compared with administration of 28% oxygen. PaCO 2 before and during oxygen administration correlated significantly (p 0.001) in both groups. Patients breathing 28% oxygen experienced a PaCO 2 fall; on the contrary, patients who received 100% oxygen showed an increase in PaCO 2, particularly those with PaCO 2 before oxygen treatment > 40 mm Hg. Conclusions: This study confirmed previous observations that oxygen dose should be variable and based on achieving and maintaining target arterial oxygen saturation measured by pulse oximetry > 92% rather than on prescribing predetermined concentrations or flow rates of inspired oxygen. (CHEST 2003; 124:1312 1317) Key words: acute severe asthma; emergency department treatment; gas exchange; hyperoxia; oxygen; oxygen therapy; uncontrolled oxygen administration Abbreviations: CI confidence interval; PEFR peak expiratory flow rate; Spo 2 pulse oximetric saturation; V /Q ventilation/perfusion Asthma exacerbations are potentially life threatening. Studies of near-fatal asthma suggest that hypoxemia is probably an important cause of death 1,2 ; therefore, oxygen supplementation is a critical part of the management of patients with acute asthma. 3 5 Because hypoxemia is produced by ventilation/perfusion (V /Q ) mismatch, mostly, it can be corrected by the administration of moderate doses of *From the Departamento de Emergencia (Drs. G. Rodrigo and Peregalli), Hospital Central de las Fuerzas Armadas, Montevideo; Centro de Tratamiento Intensivo (Dr. Rodriguez Verde), Hospital de Paysandú, Paysandú; and Unidad de Cuidado Intensivo (Dr. C. Rodrigo), Asociación Española 1 a de Socorros Mutuos, Montevideo, Uruguay. Manuscript received November 12, 2002; revision accepted March 3, 2003. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (e-mail: permissions@chestnet.org). Correspondence to: Gustavo J. Rodrigo, MD, Departamento de Emergencia, Hospital Central de las Fuerzas Armadas. Av. 8 de Octubre 3020, Montevideo 11600, Uruguay; e-mail: gurodrig@ adinet.com.uy inspired oxygen. 6,7 In spite of this, the use of highflow oxygen by a mask with reservoir bag has been assumed to be harmless and the best way to deliver it to patients with acute asthma. 8 11 Uncontrolled oxygen has been postulated to correct the effects of hypoxemia and to compensate any trend for Pao 2 to fall with -agonist therapy. Nevertheless, there is rising evidence that hyperoxia may be unsafe for some patients. One noncontrolled study 12 suggests that Paco 2 may worsen in some patients receiving 100% oxygen, especially those with more severe airflow obstruction; however, to our knowledge, there are no controlled trials that evaluate the use of oxygen in acute asthma. 13,14 In fact, the Global Strategy for Asthma Management and Prevention report 15 on asthma states that these data need to be validated in a controlled trial, but for now suggest that oxygen therapy should be titrated according to oximetry. Consequently, we conducted a randomized controlled study to assess the effect of admin- 1312 Clinical Investigations

istration of two oxygen concentrations on gas exchange in adult patients with acute severe asthma. Subjects Materials and Methods We recruited all adult patients with acute asthma who were seen in two emergency departments of Uruguay (Hospital Central de las Fuerzas Armadas in Montevideo, and Hospital de Paysandú, in Paysandú) over a 6-month period (March to August 2002). The inclusion/exclusion criteria for patients were as follows: (1) diagnosis criteria of asthma of the American Thoracic Society 16 ; (2) age from 18 to 50 years; (3) peak expiratory flow rate (PEFR) 60% of predicted value; and (4) patients were excluded if they had a temperature 38 C, or a history of cardiac, hepatic, renal disease, or other medical disease, or pregnancy; and (5) an expressed willingness to participate in the study, with written informed consent obtained. The Hospitals Ethics Committees approved the study. Design After baseline measures (the investigators made all measures unaware of the patient group assignments), subjects were randomized (using a computer-generated method) to receive one of two treatments arms. Those in the first treatment arm (28% group) received 28% oxygen for 20 min via a standard facemask (model 1088; Hudson RCI; Temecula, CA). Those in the second treatment arm (100% group) received 100% oxygen for 20 min via a standard nonrebreathing facemask (model 1060; Hudson RCI). Because facemasks used are obvious different (eg, a reservoir bag is only present in the nonrebreathing facemask), the study was not blinded. At the end of oxygen protocol, all patients received albuterol and ipratropium bromide (120 g of albuterol sulfate and 21 g of ipratropium bromide per actuation) delivered by a metered-dose inhaler into a spacer device (Volumatic; Allen & Hanburys; Greenford, UK) in a dose of four puffs at 10-min intervals in accordance with previous evidence. 17,18 Additionally, patients with a poor response received hydrocortisone, 400 mg IV. Measures The following variables were measured in each patient immediately before starting oxygen administration and during the oxygen treatment (at 20 min): PEFR, respiratory rate, heart rate, and arterial blood gases. At baseline, we also assessed the presence of accessory muscle use, dyspnea, and wheezing. PEFR was measured with a mini-wright peak flowmeter (Clement Clarke; Harlow, UK). The highest of three values was recorded. Heart rate was measured from continuous ECG. Accessory muscle use was defined as visible retraction of the sternocleidomastoid muscles. 19 Dyspnea was defined as the patient s own assessment of breathlessness. Wheezing was defined as musical or whistling breath sounds heard with a stethoscope during expiration. These clinical factors were graded in a scale from 0 to 3, in which 0 denoted absent, 1 indicated mild, 2 indicated moderate, and 3 indicated severe. Arterial blood samples were obtained via puncture of the radial artery; and ph, Pao 2, and Paco 2 were measured with a blood gas analyzer using routine techniques (ABL 500 system; Radiometer America; Westlake, OH). Arterial saturation (pulse oximetric saturation [Spo 2 ]) was monitored during oxygen administration by pulse oximetry with a finger oximeter (Nellcor N-180; Nellcor; Hayward, CA). Primary outcome measure was Paco 2 (difference between the two groups and variation from baseline) at the end of oxygen administration. Secondary outcomes variables were ph, PEFR, Pao 2, heart rate, and respiratory rate (differences between groups and variation from baseline) at the end point. Patients unable to maintain a Spo 2 90% and/or presented clinical deterioration during the oxygen protocol were excluded and received inhaled bronchodilators and systemic corticosteroids. Statistical Analysis The study was designed a priori with a sample size of 36 subjects per group to have 80% power to detect ( 0.05, two-tailed test) a mean difference between groups in Paco 2 of 2.4 mm Hg, with a SD of 5 mm Hg. Mean values SD were calculated for continuous variables. The 95% confidence intervals (CIs) were calculated with standard formulas. 20 The two groups were compared with respect to continuous variables using a repeated-measures two-way analysis of variance; post hoc pairwise multiple comparisons were performed using the Scheffé method. Pearson 2 with Yates correction or the Fisher exact test was used for categorical variables. All data were analyzed with SPSS 10.0 for Windows software (SPSS; Chicago, IL). Results A total of 79 consecutive adults with an acute exacerbation of asthma were recruited for entry into this study. Two patients were excluded because they were unable to maintain a Spo 2 90%. Seventyseven adults were randomly assigned to treatment: 38 patients to the 28% group and 39 patients to the 100% group. Three patients (two patients in the 28% group, and one patient in the 100% group) refused the use of the facemask. Finally, 74 patients (36 patients in the 28% group, and 38 patients in the 100% group) completed the study (mean age, 37.9 9.7 years). They had severe airway obstruction (mean PEFR, 41.0 12.1% of predicted), moderate hypoxemia (mean Pao 2, 77.8 12.9 mm Hg), hypocarbia (mean Paco 2, 36.4 4.4 mm Hg), and normal ph (7.39 0.02). Eight percent of patients (n 6) showed normoxemia and normocarbia; 76% (n 56) presented hypoxemia or normoxemia (range, 58 to 97 mm Hg) and hypocarbia (range, 28 to 39 mm Hg); and 16% of patients (n 12) presented hypoxemia (range, 63 to 90 mm Hg) and hypercarbia (range, 40 to 46 mm Hg). Table 1 shown baseline demographic data. There were no significant differences between the two groups, except for heart rate. Table 2 presents data at the end of oxygen administration (20 min). There were no differences between groups in heart and respiratory rates. Heart rate showed a trend toward a decrease, and respiratory rate was stable in both groups; however, patients treated with 100% oxygen presented a statistically significant decrease in ph (mean difference, 0.03; www.chestjournal.org CHEST / 124 / 4/ OCTOBER, 2003 1313

Table 1 Demographic and Baseline Characteristics of Patients* Characteristics 28% O 2 (n 36) 100% O 2 (n 38) p Value Mean Difference (95% CI) Age, yr 38.2 8.8 36.3 10.4 0.2 1.9 ( 2.5 to 6.3) Sex, male/female 24/12 (67/33) 24/14 (63/37) 0.8 Dyspnea 26 (72) 30 (79) 0.6 0.0 ( 0.3 to 0.1) Wheezing 34 (94) 38 (100) 0.3 0.0 ( 0.2 to 0.0) Accessory muscle use 27 (75) 30 (79) 0.8 0.0 ( 0.2 to 0.1) Respiratory rate, breaths/min 22.3 2.9 22.1 5.0 0.8 0.2 ( 1.7 to 2.1) Heart rate, beats/min 94.2 12.9 102.9 16.3 0.01 8.6 ( 15.5 to 1.7) Predicted PEFR, L/min 515.4 73.4 530.1 68.5 0.3 14.7 ( 47.6 to 18.1) PEFR, L/min 200.8 59.7 221.5 58.4 0.2 20.7 (13.7 to 48.5) PEFR, % predicted 38.7 11.1 43.1 12.7 0.2 4.3 ( 9.8 to 1.3) ph 7.39 0.03 7.39 0.02 0.8 0.0 ( 0.0 to 0.0) Paco 2,mmHg 36.7 3.8 36.2 4.9 0.6 0.4 ( 1.6 to 2.6) Pao 2,mmHg 79.4 11.4 76.3 14.1 0.3 3.1 ( 2.8 to 9.1) Spo 2,% 95.2 2.0 94.5 2.4 0.2 0.7 ( 0.3 to 1.8) *Data are presented as mean 1 SD or No. (%) unless otherwise indicated. 95% CI, 0.00 to 0.05) and a significant increase in Paco 2 (mean difference, 2.7 mm Hg; 95% CI, 4.7 to 0.7 mm Hg) associated with a significant decrease in PEFR (mean difference, 31.7 L/min; 95% CI, 16.7 to 46.8 L/min) compared with patients treated with 28% oxygen. As anticipated, patients who received 100% oxygen showed a substantial increase in Pao 2 compared with patients treated with 28% oxygen (mean difference, 158.4 mm Hg; 95% CI, 185.1 to 131.8 mm Hg). Figure 1 shows the relationship between Paco 2 before and at 20 min of oxygen administration. The scatterplot revealed that this relationship is curvilinear, particularly in patients treated with 100% oxygen. We tested different curve estimation regression models, and a polynomial model was the best fit to the data. Both variables correlated significantly (r 0.53, p 0.01 in the 28% group, and r 0.88, p 0.001 in the 100% group). A close examination of the relationship showed two different responses to oxygen administration. When patients received 28% oxygen, Paco 2 had a tendency to fall; on the contrary, patients who received 100% oxygen showed an increase in Paco 2, particularly in those patients with Paco 2 before oxygen treatment 40 mm Hg. Post hoc analysis of Paco 2 changes from baseline during oxygen administration are shown in Table 3. Twenty patients (55.6%) in the 28% group and 12 patients (31.6%) in the 100% group had Paco 2 decreases (p 0.05). Eight patients (22.2%) in the 28% group and eight patients (21.0%) in the 100% group (p 0.9) presented small Paco 2 increases ( 2 mm Hg) consistent with the physiologic Haldane effect. 21 Finally, only 6 patients (16.6%) in the 28% group presented a Paco 2 increase 2 mm Hg (mean increase, 3.4 1.9 mm Hg; range, 2 to 5.9 mm Hg), compared with 16 patients (42.1%) [mean increase, 5.0 3.8 mm Hg; range, 2.4 to 14.3 mm Hg] in the 100% group (p 0.02). At the end of the protocol, this subgroup of patients showed a significant differ- Table 2 Heart and Respiratory Rates, Pulmonary Function, and Arterial Blood Gas Levels at 20 min of Oxygen Administration* Variables 28% O 2 (n 36) 100% O 2 (n 38) p Value Mean Difference (95% CI) Respiratory rate, breaths/min 20.6 3.1 21.0 3.6 0.6 0.4 ( 2.1 to 1.2) Respiratory rate variation, breaths/min 1.0 2.0 1.2 2.9 0.4 0.2 ( 0.9 to 1.5) Heart rate, beats/min 92.2 13.8 96.9 17.0 0.2 4.7 (11.8 to 2.5) Heart rate variation, beats/min 2.0 6.4 4.9 8.4 0.1 2.9 ( 0.6 to 6.3) PEFR, L/min 219.7 80.1 217.8 61.9 0.8 2.6 ( 30.2 to 35.5) PEFR, variation, L/min 24.1 38.6 7.6 23.7 0.001 31.7 (16.7 to 46.8) ph 7.41 0.04 7.38 0.04 0.01 0.03 (0.00 to 0.05) ph variation 0.00 0.04 0.01 0.04 0.04 0.01 (0.00 to 0.04) Paco 2,mmHg 35.4 4.4 38.0 7.0 0.03 2.7 ( 4.7 to 0.7) Paco 2 variation, mm Hg 1.3 4.0 1.8 3.9 0.001 3.1 ( 4.9 to 1.3) Pao 2,mmHg 99.4 16.3 257.8 76.6 0.001 158.4 ( 185.1 to 131.8) *Data are presented as mean 1 SD unless otherwise indicated. 1314 Clinical Investigations

Figure 1. Paco 2 during oxygen administration as a function of Paco 2 before oxygen treatment. The variables correlated significantly in both groups (p 0.01). Patients breathing 28% oxygen had a Paco 2 fall (left panel); on the contrary, patients who received 100% oxygen showed an increase in Paco 2, particularly those with Paco 2 before oxygen treatment 40 mm Hg (right panel). ence in PEFR (PEFR deterioration of 11.9 30.0 L/min in the 100% group, compared with an increase of 18.3 25.4 L/min in the 28% group, p 0.03). Whereas Paco 2 did not correlate with PEFR in the 28% group (r 0.21, p 0.6), the 100% group showed an inverse curvilinear correlation between PEFR before oxygen administration and Paco 2 during oxygen treatment (Fig 2). Low values of PEFR were associated with high levels of Paco 2 (r 0.53, p 0.01, polynomial model). Discussion To our knowledge, this is the first randomized controlled trial that provided data on the effect of two oxygen concentrations on Paco 2 and PEFR in adult patients with acute severe asthma. The results showed that 100% of inspired oxygen resulted in a significant increase in Paco 2. Forty-two percent of the 100% oxygen group patients had increases in Paco 2 averaging 5.0 mm Hg (range, 2.4 to 14.3 mm Hg). As shown in Figures 1 and 2, Paco 2 during oxygen administration highly correlated with the Paco 2 level before oxygen treatment and moderately with the baseline airway obstruction (PEFR). Thus, baseline Paco 2 and PEFR 40 mm Hg and 35% of predicted respectively appears to be significant factors in the development of hyperoxic-induced hypercapnia. Additionally, the gas exchange deterioration was associated with a significant decrease in pulmonary function. On the contrary, the administration of 28% oxygen overall improved the patient condition. Thus, low oxygen concentration rectified hypoxemia (mean Pao 2 at the end of protocol, 99.4 16.3 mm Hg), produced a small mean Paco 2 decrease (1.3 4.0 mm Hg), and improved pulmonary function (24.1 38.6 L/min). In spite of this, 16.6% of patients experienced an increase in Paco 2 (range, 2.0 to 5.9 mm Hg). These findings are in accordance with previous studies. More than 10 years ago, two noncontrolled trials 22,23 found that acute asthma patients treated with high inspired oxygen concentration presented a significant increase in Paco 2 (of the order of 3 to 5 mm Hg). More recently, Chien et al 12 demonstrated that the administration of 100% oxygen to acute asthma patients with moderate-to-severe airway obstruction (FEV 1 of 49.1% of predicted) may adversely influence carbon dioxide elimination. With 100% oxygen administration, FEV 1 fell by 4.3%, ph Table 3 Distribution of PaCO 2 Changes From Baseline During Oxygen Administration* Paco 2 Change 28% O 2 (n 36) 100% O 2 (n 38) p Value Mean Difference (95% CI) No change 2 (5.6) 2 (5.3) 0.9 0.00 ( 0.12 to 0.13) Fall 1 mm Hg 20 (55.6) 12 (31.6) 0.05 0.24 (0.02 to 0.43) Rise 2 mm Hg (Haldane effect) 8 (22.2) 8 (21.0) 0.9 0.01 ( 0.17 to 0.19) Rise 2 mm Hg 6 (16.6) 16 (42.1) 0.02 0.25 ( 0.43 to 0.05) *Data are presented as No. (%) unless otherwise indicated. www.chestjournal.org CHEST / 124 / 4/ OCTOBER, 2003 1315

Figure 2. Relationship between airway obstruction before oxygen administration as measured by PEFR (as percentage of predicted) and Paco 2 during the administration of 100% oxygen. Low values of PEFR were associated with high levels of Paco 2 (r 0.54, p 0.01). decreased by 0.02, and Paco 2 rose by 2.3 mm Hg). Almost 41% of patients had elevations in Paco 2 2 mm Hg, averaging 5.9 mm Hg. In general, the tendency toward hypercarbia was the greatest in the patients with the most abnormal baseline condition (FEV 1 and Paco 2 ). Although our study did not look closely at any changes in minute ventilation, there seemed to be no significant modifications before and during oxygen administration in respiratory rate. Hyperoxia associated with hypercarbia occurring in asthma exacerbations, and especially without any evidence of respiratory suppression, would be more easily explained by the regional release of hypoxic pulmonary vasoconstriction. This factor had a major role in determining V /Q matching, as V /Q inequality worsened considerably after administration of 100% oxygen. This has been shown in patients with acute asthma receiving ventilation and not receiving ventilation 22,23 ; therefore, patients with most severe baseline condition probably have more hypoxic vasoconstriction, and they had the greatest increase in Paco 2 while breathing 100% oxygen. Our trial sample presented the typical features of severe adult asthmatic patients when they presented for care to an emergency department on average: a mean level of PEFR of 41% of predicted, a mean age of 38 years, and a female/male ratio of 2:1. Finally, 76% of our patients presented the most characteristic arterial blood gas pattern: mild-to-moderate hypoxemia (range, 58 to 97 mm Hg) along with hypocapnia (range, 28 to 39 mm Hg). 24,25 In summary in this randomized, controlled trial, we have confirmed previous observations that administration of 100% oxygen for 20 min significantly increases Paco 2 and decreases PEFR as compared with administration of 28% oxygen. These observations strongly support Global Strategy for Asthma Management and Prevention recommendations 15 that oxygen dose in severe acute asthma should be variable and should be based on achieving and maintaining target Spo 2 values with a pulse oximeter 92% for adults 26 and 95% for children. Because severe asthma may give rise to hypoxemia and -agonist therapy may worsen hypoxemia, monitoring therapy of acute asthma with pulse oximetry is ideal. 27 The results should be used to control the administration and dose of oxygen therapy. In circumstances where acute asthma must be treated without pulse oximetry, a clinical judgment must be made as to whether oxygen therapy (if is available) is warranted to prevent life-threatening hypoxemia, despite its potential adverse effects. However, uncontrolled high-flow oxygen should be avoided. References 1 Molfino NA, Nannini LJ, Martelli AN, et al. Respiratory arrest in near-fatal asthma. N Engl J Med 1991; 324:285 288 2 Molfino NA. Near-fatal asthma. In: Hall JB, Corbridge T, Rodrigo C, et al, eds. Acute asthma: assessment and management. New York, NY: McGraw-Hill, 2000; 29 47 3 National Asthma Education and Prevention Program. Expert panel report 2: Guidelines for the diagnosis and management of asthma. Bethesda, MD. National Institutes of Health, 1997; publication No. 55 4051 4 British guidelines on asthma management 1995: review and position statement. Thorax 1997; 53(Suppl):S1 S24 5 Canadian asthma consensus report. Management of patients with asthma in the emergency department and in hospital. Can Med Assoc J 1999; 161(Suppl):S53 S59 6 Rodriguez-Roisin R. Acute severe asthma: pathophysiology and pathobiology of gas exchange abnormalities. Eur Respir J 1997; 10:1359 1371 7 Rodriguez-Roisin R. Gas exchange in acute asthma. In: Hall JB, Corbridge T, Rodrigo C, et al, eds. Acute asthma: assessment and management. New York, NY: McGraw-Hill, 2000; 83 103 8 British Thoracic Society and others. Guidelines for the management of asthma: management of acute asthma. Thorax 2003; 58:i32 i50 9 Scottish Intercollegiate Guidelines Network. Emergency management of acute asthma: a national clinical guideline. Edinburgh, Scotland: Scottish Intercollegiate Guidelines Network, 1999; Sign Publication No. 38 10 Inwald D, Roland M, Kuitert S, et al. Oxygen treatment for acute severe asthma. BMJ 2001; 323:98 100 11 Werner HA. Status asthmaticus in children: a review. Chest 2001; 119:1913 1929 12 Chien JW, Ciufo R, Novak R, et al. Uncontrolled oxygen administration and respiratory failure in acute asthma. Chest 2000; 117:728 733 1316 Clinical Investigations

13 FitzGerald M. Extracts from clinical evidence : acute asthma. BMJ 2001; 323:841 845 14 Thomson AJ, Webb DJ, Maxwell SR. Oxygen therapy in acute medical care: the potential dangers of hyperoxia need to be recognised. BMJ 2002; 321:1406 1407 15 NHLBI/WHO workshop report. Global strategy for asthma management and prevention. Bethesda, MD: National Institutes of Health, 2002; Publication No. 02 3659 16 American Thoracic Society. Standards for the diagnosis and care of patients with chronic obstructive pulmonary disease (COPD) and asthma. Am Rev Respir Dis 1987; 136:225 244 17 Rodrigo C, Rodrigo G. Therapeutic response patterns to high and cumulative doses of salbutamol in acute severe asthma. Chest 1998; 113:593 598 18 Rodrigo GJ, Rodrigo C. The role of anticholinergics in acute asthma treatment: an evidence-based evaluation. Chest 2002; 121:1977 1987 19 McFadden ER, Kiser R, degroot WJ. Acute bronchial asthma: relations between clinical and physiologic manifestations. N Engl J Med 1973; 288:221 225 20 Altman DG, Machin D, Bryant TN, et al. Statistics with confidence. Bristol, UK: BMJ Books, 2000 21 Lenfant C. Arterial-alveolar difference in Pco 2 during air and oxygen breathing. J Appl Physiol 1966; 21:1356 1362 22 Rodriguez-Roisin R, Ballester E, Torres A, et al. Mechanisms of abnormal gas exchange in patients with status asthmaticus needing mechanical ventilation. Am Rev Respir Dis 1989; 139:732 739 23 Ballester E, Reyes R, Roca J, et al. Ventilation-perfusion mismatching in acute severe asthma: effects of salbutamol and 100% oxygen. Thorax 1989; 44:258 267 24 Tai E, Read J. Arterial-blood gas tension in asthma. Lancet 1967; 1:644 646 25 McFadden ER, Lyons H. Arterial blood gas tension in asthma. N Engl J Med 1968; 278:1027 1032 26 Carruthers DM, Harrison BDW. Arterial blood gas analysis or oxygen saturation in the assessment of acute asthma? Thorax 1995; 50:186 188 27 Rodrigo GJ. Oxygen treatment for acute severe asthma: oxygen saturation may help identify patients in need of intensive management [letter]. BMJ 2001; 323:1069 www.chestjournal.org CHEST / 124 / 4/ OCTOBER, 2003 1317