Pneumoconiosis

Study objective: To investigate the clinical features and prognosis of patients with coal worker's pneumoconiosis (CWP) requiring invasive mechanical ventilation (MV) in the ICU for their first episode of acute respiratory failure (ARF), with special attention to the prognostic implication of radiographic progressive massive fibrosis (PMF).

Design: Retrospective study.

Setting: A 16-bed medical ICU at a community hospital.

Patients and methods: We reviewed 53 patients with CWP and ARF requiring invasive MV in the ICU for the first time between August 1998 and March 2002.

Results: Of the 53 patients with CWP, 28 patients (53%) with PMF had their first ARF at a younger age than those without PMF (69.1 [+ or -] 7.9 years vs 74.8 [+ or -] 7.2 years, p = 0.008 [mean [+ or -] SD]). Pneumonia (49%) was the most common cause of ARF. The mean APACHE (acute physiology and chronic health evaluation) II score was 26.0 [+ or -] 9.9, and the mean ICU stay was 14.7 [+ or -] 16.1 days. Twenty-one patients (40%) were weaned successfully in the ICU, with mean ventilator time of 17.1) [+ or -] 25.1 days. The ICU and in-hospital mortality rates were 40% and 43%, respectively. The median survivals for all patients and the ICU survivors were 2.6 months and 14.3 months, respectively. Multivariate analysis showed the following risk (or protective) factors for the ICU mortality: Pa[CO.sub.2] > 45 mm Hg at the time of intubation (adjusted odds ratio [OR], 0.04; 95% confidence interval [CI], 0.003 to 0.44), Pa[O.sub.2]/fraction of inspired oxygen ratio < 200 mm Hg at the time of intubation (OR, 8.78; 95% CI, 1.36 to 56.48), and APACHE II score [greater than or equal to] 25 (OR, 11.99; 95% CI, 1.49 to 96.78). PMF was not associated with the ICU mortality (OR, 1.18; 95% CI, 0.20 to 7.10).

Conclusions: Radiographic PMF was not associated with the ICU mortality in patients with CWP and ARF receiving invasive MV in the ICU. Although a substantial proportion of them could be weaned from the ventilator and discharged from the hospital, their long-term prognosis was poor.

Key words: acute respiratory failure: coal worker's pneumoconiosis; mechanical ventilation

Abbreviations: ABG arterial blood gas; APACHE = acute physiology and chronic health evaluation; ARF = acute respiratory failure; CI = confidence interval; CWP = coal worker's pneumoconiosis; FI[O.sub.2] = fraction of inspired oxygen; MV = mechanical ventilation: OR = odds ratio; PMF = progressive massive fibrosis

**********

The outcome data from acute respiratory failure (ARF) are vital in assessing the immediate risk of death of a critically ill patient, in predicting the long-term prognosis and pulmonary functional status, as well as in making clinical decisions. There have been many studies looking at the outcomes of ARF in various pulmonary diseases, including asthma, (1) severe pneumonia, (2) ARDS, (3) COPD, (4-10) interstitial lung diseases, (11-13) and lung cancer. (14) The worst outcomes were noted in patients with lung cancer and pulmonary fibrosis requiting invasive mechanical ventilation (MV) for ARF, with hospital mortality rates of 91.3% and nearly 100%, respectively. (11-14) In a study of patients with chronic respiratory insufficiency receiving domiciliary oxygen therapy or MV, the outcome of patients with pneumoconiosis is the worst, similar to those with pulmonary fibrosis, with 50% dying during the year following the beginning of home treatment. (15)

The pattern and severity of respiratory impairment in coal worker's pneumoconiosis (CWP) are related to the levels of coal mine dust exposure, geologic factors, exposure to other respiratory hazards, and immunologic response to dusts. (36-18) Estimation of disease severity in CWP conventionally involves a combination of radiographic and lung function assessment. (17,18) Radiographically, CWP is classified into two forms: simple and complicated, based on the absence or presence of progressive massive fibrosis (PMF). (17,18) In patients with simple CWP, lung function is minimally impaired and life expectancy is similar to that of the general population. (16-18) However, prominent obstructive and restrictive abnormalities have been noted in patients with complicated CWP, whose survival rates are reduced substantially. (16-18) This does not imply that the outcome of patients with CWP and ARF would be worse ill complicated CWP, because clinical observations have shown that considerable variations in the magnitude of pulmonary impairment exist in complicated CWP. (19) Since there is a lack of outcome data on patients with CWP and ARF, we undertook this study to investigate this issue and the prognostic implications of PMF.

MATERIALS AND METHODS

Patients

The purpose of tiffs retrospective study was to investigate the outcomes of patients with CWP who acquired ARF and received invasive MV for the first time. This study was carried out in a northern Taiwan hospital serving a community with > 90 years of coal mining history and clusters of pottery manufacturers. The coal mined was distributed in an area of 2,400 [km.sup.2] containing largely medium volatile bituminous and subbituminous coal. During the study period front August 1997 to March 2002, there were 1,961 ex-coal miners who visited this hospital, and 530 patients with a diagnosis of pneumoconiosis who were hospitalized due to various disorders. We analyzed the medical records of patients treated in a 16-bed medical ICU of the hospital. Case records were selected through a computerized search of diagnoses codified at hospital discharge. The diagnosis of CWP was based on both a history of prolonged exposure to coal dust and radiographic evidence compatible with pneumoconiosis according to the International Labor Office classification system. (20) Patients were excluded if they were not coal miners, were intubated for nonmedical reasons, were transferred to another hospital, or if they had prior intubations for ARF. Patients were also excluded if they were only treated with noninvasive MV because they might be not transferred to the ICU. The requirement for Human Subjects Institutional Review Board approval was waived due to the low risk of this study to the patients.

For each patient, we reviewed all available pertinent chest radiographs within 3 years before admission to the hospital. The chest radiographs obtained at the nearest time before intubations were selected for radiographic classification. The images were reinterpreted and compared with standard reference radiographs by two pulmonologists who did not know the patients' clinical conditions. In case they disagreed, a consensus was made. The designation of radiographic profusion category was simplified as a three level classification (1 = 0/-, 0/0, 0/1, 1/0, 1/1, 1/2; 2 = 2/1, 2/2, 2/3; 3 = 3/2, 3/3, 3/+). Small opacities in the parenchyma were classified by shape and size: p, q, or r for rounded opacities (diameter, < 1.5 mm, 1.5 to 3 mm, or > 3 mm, respectively) and s, t, or u for irregular opacities (of width, < 1.5 mm, 1.5 to 3 mm, or > 3 mm, respectively). (20) Radiographic PMF was defined by tire presence of one or more large opacities [greater than or equal to] 1 cm in diameter, and was classified as A (for one or more such opacities but not exceeding a combined diameter of 5 cm), B (one or more larger opacities than A and whose combined area does not exceed the upper zone of the right lung), or C (larger than B). (20) In cases with poor radiograph quality or a standing radiograph not available, only the presence of PMF was addressed. For small opacities, only the predominant types were recorded.

Data Collection and Definitions

The following clinical data were collected for each patient: age and sex; body mass index; baseline pulmonary functions obtained by body plethysmography (V6200 Autobox; SensorMedics; Yorba Linda, CA); smoking, mining and medical histories; clinical symptoms; ECG for cor pulmonale; (4) APACHE (acute physiology and chronic health evaluation) II score; serum albumin and arterial blood gas (ABG) levels on ICU admission; causes of ARF; durations of invasive MV before hospital discharge and ICU stay; and outcome.

A positive smoking history was designated for either current or ex-smokers. Chronic respiratory symptoms were defined as chronic cough for > 3 months with or without dyspnea, and with or without long term use of bronchodilators (> 3 months). Cor pulmonale was defined by at least one of the following three ECG criteria: presence of p pulmonale, right ventricular hypertrophy, or right-axis deviation. (4) Definitions of various medical diseases were as follows: tuberculosis for those with documented or treated cases; diabetes mellitus for those with a fasting glucose level > 126 mg/dL, or a history of prior use of insulin or oral bypoglycemic agents; hypertension for those with documented history. (systolic BP [greater than or equal to] 140 mm Hg or diastolic BP [greater than or equal to] 90 mm Hg on more than two clinic visits), with or without antihypertensive treatment; coronary artery diseases for those with a history of myocardial infarction or confirmed by coronary angiography. Pneumonia was defined as the presence of fever, new or aggravated productive cough, dyspnea, or leukocytosis (WBC count > 11,000/[micro]L), and new pulmonary infiltrates on the chest radiograph. Acute exacerbation of chronic respiratory insufficiency was considered if isolated respiratory tract infection without new pulmonary infiltrates was diagnosed, or the cause of ARF could not be identified. Shock was defined as systolic BP < 90 mm Hg, which lasted for > 4 h or required vasopressor use.

Those who were readmitted to the ICU within 72 h after being discharged were considered to have the same ICU stay; otherwise, the ICU readmission was regarded as a hospital stay. All patients received MV with the Evita 2 ventilators (Dragerwerk; Aktieugesellschaft, Germany). Weaning failure was defined by the inability to maintain spontaneous breathing for > 72 h, or death within 72 h after extubation. For patients discharged alive from the hospital, the outcomes concerning survival and ventilator dependency were confirmed by follow-up telephone interviews and death certificates. For patients unable to be contacted, the latest dates of known follow-up during the study period were designated as the censor dates.

Statistics

Statistical analysis was performed with a computer program (SPSS for Windows, version 10.0; SPSS; Chicago, IL). Values were expressed as means [+ or -] SD. Comparisons between those with and without PMF and between ICU survivors and nonsurvivors were made using the Student t test, and [chi square] or Fisher exact test. The Pearson's product-moment correlation was used to assess the relationship between the two continuous variables. To investigate the prognostic factors, proportions of variables in the univariate analysis were compared with the [chi square] test after excluding variables with missing values. Multivariate analysis was performed with entering all the variables in the logistic regression model. To assess and present the potential confounding factors of the ICU mortality more clearly, both the crude and adjusted odds ratios (ORs) are calculated. A p value of < 0.05 was considered significant. A life table was computed using the Kaplan-Meier method.

RESULTS

During the study period, a total of 72 patients with pneumoconiosis and ARF were admitted to the ICU. Nineteen of them were excluded, including 4 pottery workers, 1 patient who was admitted due to a suicide attempt, 7 patients who were intubated for elective operations, 4 patients with previous intubation for ARF, 2 patients who did not receive endotracheal intubation, and 1 patient who was transferred to another hospital before weaning. Of the remaining 53 patients, 2 patients were intubated after a failed noninvasive MV, and 3 patients were readmitted to the ICU for invasive MV within 72 h after extubation. The clinical features, and durations of MV and ICU stay are summarized in Table 1. The ICU survivors had more frequent cor pulmonale on ECG (10 of 32 patients vs 1 of 21 patients, p = 0.035), higher Pa[CO.sub.2] levels at the time of intubation (51.9 [+ or -] 22.0 mm Hg vs 39.4 [+ or -] 19.2 mm Hg, p = 0.034), a lower APACHE 11 score (22.6 [+ or -] 7.5 vs 31.3 [+ or -] 10.9, p = 0.003), and longer duration of MV (33.0 [+ or -] 39.7 days vs 9.5 [+ or -] 14.6 days, p = 0.004). There was a trend toward worse oxygenation in the ICU nonsurvivors (Pa[O.sub.2]/fraction of inspired oxygen [FI[O.sub.2]], 266.4 [+ or -] 149.5 mm Hg vs 200.1 [+ or -] 120.2 mm Hg, p = 0.081).

Radiographic classification of pneumoconiosis was performed in 44 patients (83%) with available standard standing radiographs. Of the nine noncategorized patients, five patients were shown to have PMF. Comparisons between patients with simple and complicated CWP are summarized in Table 2. Patients with complicated CWP had a younger age, more chronic respiratory symptoms, poorer lung function, and a higher Pa[CO.sub.2] level at the time of intubation. In addition, the radiographic classification for those with complicated CWP showed that they belonged to the more severe category (profusion category 3), and they had more type u in small opacities. Among the causes of ARF, there was a trend for more acute exacerbations of chronic respiratory insufficiency in those with PMF (Table 2). However, the Pa[O.sub.2]/FI[O.sub.2] ratio, APACHE II score, and the ICU mortality rates were similar.

The causes of ARF included pneumonia (n = 26), acute exacerbation of chronic respiratory insufficiency (n = 13), massive hemoptysis (n = 3), sputum impaction (n = 3), pneumothorax (n = 2), acute myocardial infarction (n = 2), theophylline intoxication (n = 1), sepsis caused by cellulites (n = 1), hypovolemic shock caused by GI bleeding (n = 1), and ventricular arrhythmia (n = 1). Cardiopulmonary resuscitation was performed in five patients at the time of intubation; all of them survived for > 24 h, but only one patient who was intubated due to tension pneumothorax survived to hospital discharge. Shock was present in eight patients on ICU admission; four of them died in the ICU.

Twenty-one patients (40%), including 4 patients undergoing tracheostomy, were weaned successfully in the ICU with a ventilator time of 17.0 [+ or -] 25.1 days; all of them survived to hospital discharge. Tracheostomy was performed in 11 patients (21%) with a mean time of 21.6 [+ or -] 16.8 days after intubation; 9 of these patients survived to hospital discharge. Of the 11 patients who were not weaned during the ICU stay, 2 died in the hospital, 7 became ventilator dependent, and 2 were weaned eventually but were bedridden and required continuous oxygen support. Of the seven miners who survived but remained ventilator dependent, two were eventually weaned from ventilator (5 months and 7 months after discharge, respectively), one was still ventilator dependent 5 months after discharge, three died (1 month, 2 months, and 6 months after discharge, respectively), and one was unavailable for follow-up. The ICU and in-hospital mortality rates were 40% and 43%, respectively. The survival curve of the 53 patients is plotted in Figure 1. Two patients were unavailable for follow-up after discharge. The median survivals for all patients and the 32 ICU survivors were 2.6 months and 14.3 months, respectively. There was no significant difference between the survival of the patients with simple CWP and those with complicated CWP (log-rank test, p = 0.772).

[FIGURE 1 OMITTED]

The relative influence of variables on the ICU mortality (univariate and multivariate analysis) is summarized in Table 3. There were three independent variables associated with the ICU mortality: Pa[CO.sub.2] > 45 mm Hg (adjusted OR, 0.04; 95% confidence interval [CI], 0.003 to 0.44), Pa[O.sub.2]/[FI[O.sub.2] ratio < 200 mm Hg (OR, 8.78; 95% CI, 1.86 to 56.48) at the time of intubation, and APACHE II score [greater than or equal to] 25 (OR, 11.99; 95% CI, 1.49 to 96.78). In a further comparison between patients with Pa[CO.sub.2] > 45 mm Hg (hypercapnia) and those with Pa[CO.sub.2] > 45 mm Hg at the time of intubation, patients with hypercapnia had a lower mean APACHE II score (22.7 [+ or -] 8.7 vs 28.4 [+ or -] 10.2, p = 0.035); but the correlation between Pa[CO.sub.2] and APACHE II score in all patients with CWP was poor (r = -0.220, p = 0.114).

DISCUSSION

The findings of this study show that the long-term prognosis of patients with CWP alter ARF was poor even though a substantial proportion of them could be weaned from the ventilator and discharged from the hospital. Radiographic PMF was relatively common in CWP with ARF, but its presence did not predict a poor ICU outcome.

The prognosis of patients with ARF requiring MV is different among various populations. (1-14) In a review (5) of unrestricted series of patients requiring MV for ARF, the 1-year mortality rates ranged from 60 to 84%. The worst outcomes were noted in patients with lung cancer and pulmonary fibrosis requiring invasive MV for ARF with hospital mortality rates of 91.3% and nearly 100%, respectively. (11-14) In our series, the in-hospital (43%) and 1-year (approximately 60%) mortality rates were similar to those in COPD with ARF (20 to 50% and 44 to 66%, respectively, in studies since 1980). (4-10)

In patients with COPD and ARF, there have been many reported prognostic variables, including the following: (1) features reflecting physiologic reserve, such as premorbid level of activity, FE[V.sub.1], serum albumin level, severity of dyspnea, and presence of cor pulmonale or left ventricular failure; (2) nature and severity of acute illness, such as causes of ARF, initial blood gas data, and APACHE II score; and (3) development of complications, especially nonpulmonary organ failure. (4,7,10) Due to different selection criteria and study designs, the specific variables associated with mortality are not consistent in these reports; nevertheless, mortality often is related to the nature of the precipitating illness and the severity of underlying chronic respiratory disease.

In this study, we found three independent variables were associated with the ICU mortality: Pa[CO.sub.2] and Pa[O.sub.2]/FI[O.sub.2] ratio at the time of intubation, and APACHE II score. The influence of cor pulmonale on the ICU mortality was not present after adjusting for other variables, since its presence favors COPD or other illnesses resulting in chronic respiratory insufficiency, which may also result in hypercapnia, a major prognostic factor shown in this study. It is surprising that a higher Pa[CO.sub.2] level at the time of intubation was associated with a better ICU outcome. The role of the Pa[CO.sub.2] level on the outcome of patients with chronic respiratory diseases has been controversial. (15,21-25) In COPD before long-term oxygen therapy, hypercapnia on room air was found to be a negative prognostic factor. (21) In patients with chronic respiratory insufficiency (including chronic bronchitis, asthma, bronchiectasis, tuberculosis sequelae, kyphoscoliosis, pulmonary fibrosis, and pneumoconiosis, except neuromuscular disorders) requiring domiciliary oxygen therapy or MV, hypercapnia on room air has been found to be a positive prognostic factor (15,22-24); however, some studies (15,21,25) showed a contrary result. In severe COPD with acute exacerbation with or without invasive MV, premorbid hypercapnia was not associated with survival. (4,8) Since we did not encounter premorbid levels of Pa[CO.sub.2] and the patient population we investigated was different, it is difficult to compare our data with previous studies in this aspect. Nevertheless, there are several possible reasons that may explain the better prognosis of hypercapnic patients with CWP in this study. First, the severity of acute illness in the hypercapnic patients was milder, since the mean APACHE II score was lower compared with those without hypercapnia. Second, an obstructive component of CWP might have played a role. Since most of the patients were smokers (76%), the prognostic analysis might be confounded by a greater proportion of COPD, which would account for the higher Pa[CO.sub.2] level, increased frequency of cor pulmonale, and increased frequency of acute exacerbation as the cause of ARF in the ICU survivors. By contrast, the lower Pa[O.sub.2]/FI[O.sub.2] ratio in those who died in the ICU might suggest a greater proportion of functionally significant interstitial disease in that group. Third, the blood gas was obtained at the time of intubation, and the stability may have influenced the result.

Since epidemiologic studies (16-18) have shown that patients with complicated CWP tend to have prominent obstructive and restrictive ventilatory abnormalities and reduced survival rates, it is striking that the presence of PMF was not associated with the ICU outcome in this study. There are several possible explanations. First, a selection bias is likely. Patients with more severe category of complicated CWP and poorer lung function might be not intubated and aggressively treated, and the result would be biased toward indiscriminate between simple and complicated CWP. Second, the presence of a higher Pa[CO.sub.2] level, a protective factor in this study, was observed in patients with complicated CWP. Third, the magnitude of pulmonary impairment among patients with complicated CWP might vary considerably. (19) It is well known that PMF is not always a homogeneous mass of fibrous tissue, but in many cases has a soft core with substantially reduced collagenous material, which may act only as little more than a space-occupying lesion without significant compensatory emphysema. (19) Therefore stone patients with CWP exhibiting dramatic radiologic changes might have little or no accompanying disablement. (19)

We are aware that there are other factors that might have adversely affected the findings of our study. First, analysis based on severity of pneumoconiosis and pre-existing lung disease might have been more informative, but this is not feasible because lung function data were not available on more than half of the study population. Besides, the available lung function might be biased toward higher values since only less severe patients could perform lung function tests. Second, the follow-up duration was relatively short; thus, the long-term prognosis might have been underestimated. Nevertheless, it is important to be aware that the presence of PMF, a poor radiologic appearance, is not a predictor of poor outcome in patients with CWP who are intubated for ARF, but the results need to be confirmed by further studies.

In conclusion, radiographic PMF was relatively common in patients with CWP and ABF receiving invasive MV in the ICU, but its presence was not associated with the ICU mortality. Although a substantial proportion could be successfully weaned from the ventilator and discharged from the hospital, the long-term prognosis was poor.

REFERENCES

(1) Marquette CH, Saulnier F, Leroy O, et al. Long-term prognosis of near-fatal asthma: a 6-year follow-up study of 145 asthmatic patients who underwent mechanical ventilation for a near-fatal attack of asthma. Am Rev Respir Dis 1992; 146:76-81

(2) Lerny O, Santre C, Beuscart C, et al. A five-year study of severe community-acquired pneumonia with emphasis on prognosis in patients admitted to an intensive care unit. Intensive Care Med 1995; 21:24-31

(3) Jerng JS, Yu CJ, Liaw YS, et al. Clinical spectrum of acute respiratory distress syndrome in a tertiary referral hospital: etiology, severity, clinical course, and hospital outcome. J Formos Med Assoc 2000; 99:538-543

(4) Menzies R, Gibbons W, Goldberg P. Determinants of weaning and survival among patients with COPD who require mechanical ventilation for acute respiratory failure. Chest 1989; 95:398-405

(5) Stauffer JL, Fayter NA, Graves B, et al. Survival following mechanical ventilation for acute respiratory failure in adult men. Chest 1993; 104:1222-1229

(6) Portier F, Defouilloy C, Muir JF. Determinants of immediate survival among chronic respiratory insufficiency patients admitted to art intensive care unit for acute respiratory failure: a prospective multicenter study. Chest 1992; 101:204-210

(7) Seneff MG, Wagner DP, Wagner RP, et al. Hospital and 1-year survival of patients admitted to intensive care units with acute exacerbation of chronic obstructive pulmonary disease. JAMA 1995; 274:1852-1857

(8) Connors AF, Dawson NV, Thomas C, et al, for the SUPPORT investigators. Outcomes following acute exacerbation of severe chronic obstructive lung disease. Am J Respir Crit Care Med 1996; 154:959-967

(9) Hill AT, Hopkinson RB, Stableforth DE. Ventilation in a Birmingham intensive care unit 1993-1995: outcome for patients with chronic obstructive pulmonary disease. Respir Med 1998; 92:156-161

(10) Afessa B, Morales IJ, Scallion PD, et al. Prognostic factors, clinical course, and hospital outcome of patients with chronic obstructive pulmonary disease admitted to an intensive care unit for acute respiratory failure. Crit Care Med 2002; 30:1610-1615

(11) Fumeaux T, Rothmeier C, Jolliet P. Outcome of mechanical ventilation for acute respiratory failure in patients with pulmonary fibrosis. Intensive Care Med 2001; 27:1868-1874

(12) Blivet S, Philit F, Sab JM, et al. Outcome of patients with idiopathic pulmonary fibrosis admitted to the ICU for respiratory failure. Chest 2001; 120:209-212

(13) Stern JB, Mal H, Groussard O, et al. Prognosis of patients with advanced idiopathic pulmonary fibrosis requiring mechanical ventilation for acute respiratory failure. Chest 2001; 120:213-219

(14) Ewer MS, All MK, Atta MS, et al. Outcome of lung cancer patients requiring mechanical ventilation for pulmonary failure. JAMA 1986; 256:3364-3366

(15) Chailleux E, Fauroux B, Binet F, et al. Predictors of survival in patients receiving domiciliary oxygen therapy or mechanical ventilation: a 10-year analysis of ANTADIR observatory. Chest 1996; 109:741-749

(16) Coggon D, Taylor AN. Coal mining and chronic obstructive pulmonary disease: a review of the evidence. Thorax 1998; 53:398 407

(17) Rom WN. Environmental and occupational medicine. 3rd ed. Philadelphia, PA: Lippincott-Raven, 1998

(18) Becklake MR, Cowie RL. Pneumoconioses. In: Murray JF, Nadel JA, eds. Textbook of respiratory medicine (vol 2). 3rd ed. Philadelphia PA: W.B. Saunders, 2000; 1811-1851

(19) Lyons JP, Campbell H. Relationship between progressive massive fibrosis, emphysema, and pulmonary dysfunction in coalworkers' pneumoconiosis. Br J Ind Med 1981; 38:125-129

(20) International Labour Office. Guidelines for the use of ILO international classification of radiographs of pneumoconioses. Occupational Safety Health Series 1980; 22:1-48

(21) Boushy SF, Thompson HK, North LB, et al. Prognosis in chronic obstructive disease. Am Rev Respir Dis 1973; 108: 1373-1383

(22) Noctournal Oxygen Therapy Trial Group. Continuous or nocturnal therapy in hypoxemic obstructive lung disease: a clinical trial. Ann Intern Med 1980; 93:391-398

(23) Keller R, Ragaz A, Borer P. Predictors for early mortality in patients with long-term oxygen home therapy. Respiration 1985; 48:216-221

(24) Dubois P, Jamart J, Machiels J, et al. Prognosis of severely hypoxemic patients receiving long-term oxygen therapy. Chest 1994; 105:469-474

(25) Medical Research Council Working Party. Long-term domiciliary oxygen therapy in chronic hypoxic cor pulmonale complicating chronic bronchitis and emphysema. Lancet 1981; 1:681-686

* From the Department of Internal Medicine (Dr. Shen), En-Chu-Kong Hospital: and Department of Internal Medicine (Drs. Jerng, Yu, and Yang), National Taiwan University Hospital, Taipei, Taiwan.

The work was performed in En-Chu-Kong Hospital, Taipei, Taiwan.

Manuscript received April 7, 2003; revision accepted September 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: Jih-Shuin Jerng, MD, Department of Internal Medicine, National Taiwan University Hospital, No. 7 Chung Shan South Rd, 100 Taipei, Taiwan; e-mail: jsjerng@ha.mc.ntu. edu.tw

COPYRIGHT 2004 American College of Chest Physicians
COPYRIGHT 2004 Gale Group