-
Lung Cancer
2001 Update on Diagnosis, Staging, &
Treatment
James N. Allen, M.D.
-
-
A. Epidemiology:
- 1. 170,000 new cases per year
- 2. 140,000 deaths per year
- 3. currently the #1 cause of cancer deaths in both men and
women (figure)
- 4. racial variations in incidence exist (figure)
- 5. incidence increases with age (figure)
B. Etiology (figure):
- 1. tobacco smoke - 85%
- a) risk of cancer correlates with age of smoking onset,
number of cigarettes smoked per day and duration of smoking
(figure)
- (1) "pack-year" = (# packs smoked per day) x (# years
smoked)
- b) currently 28% women and 32% of men smoke
- c) 80% of smokers start before age 18
- 2. asbestos - 5% (figure)
- 3. radon - 5% (figure)
- 4. environmental tobacco smoke - 1% (figure)
- 5. other minor risk factors:
- a) arsenic
- b) BCME & CMME (chloromethyl ethers)
- c) chromium
- d) mustard gas
- e) nickel
- f) vinyl chloride
- g) family history of lung cancer
C. Classification (figure):
- 1. small cell
- 2. non-small cell
- a) adenocarcinoma
- b) squamous cell carcinoma
- c) large cell carcinoma
- d) bronchoalveolar carcinoma
D. Clinical Manifestations:
- 1. Common signs & symptoms:
- a) anorexia
- b) cough
- c) chest pain
- d) dyspnea
- e) hemoptysis
- f) supraclavicular node enlargement
- g) bone pain
- h) neurologic manifestations
- i) hoarseness (left recurrent laryngeal nerve
involvement)
- j) post-obstructive pneumonia
- k) ** 5-15% asymptomatic at the time of diagnosis
- 2. Paraneoplastic syndromes (occur in 10-20% of patients):
- a) clubbing
- b) superficial thrombophlebitis
- c) hypercalcemia
- d) hyponatremia
- e) peripheral neuropathy
- f) Lambert-Eaton syndrome (myasthenia gravis - like
condition)
E. Diagnostic Approach:
- 1. Radiography
- a) screening chest x-rays not cost-effective and do not
affect overall mortality
- b) chest x-ray is overall 70-88% accurate in overall
detection of lung cancer
- (1) 61-71% accurate for hilar nodes
- (2) 47-60% accurate for mediastinal nodes
- c) chest CT permits identification of enlarged lymph nodes,
adrenal masses, and liver metastases involving the superior
portion of the liver
- (1) lymph nodes and adrenal massess can be benign and if
the patient is otherwise and operative candidate, nodes
and/or the adrenal glands should always be biopsied if
abnormal by CT (figure)
- 2. Sputum Cytology
- a) diagnostic in up to 20% of all patients with lung cancer
- (1) positive in < 5% of patients with peripheral
cancers
- b) false positives do occur
- c) screening detects cancer earlier but due to lead time
bias, there is no survival benefit
- 3. Bronchoscopy
- a) solitary pulmonary nodules
- (1) yield 10% if nodule is < 2 cm
- (2) yield 40-50% if nodule is >2-4 cm
- b) optimal # biopsies = 3-4
- c) overall diagnostic accuracy for visible lesions = 94%
- (1) washings = 76%
- (2) brushings = 74%
- (3) biopsy = 82%
- d) overall diagnostic accuracy for peripheral lesions (all
sizes) = 86%
- (1) washings = 52%
- (2) brushings = 52%
- (3) biopsy = 61%
- 4. Trans-Thoracic Needle Aspirate
- a) diagnostic yield is 43-97% if fluoroscopically
visible
- b) aspirate cytology correlates 65% with operative
histology
- c) complications:
- (1) pneumothorax 35%
- (2) hemoptysis 7%
- (3) chest tube placement 6%
- 5. Mediastinoscopy - often useful in establishing a diagnosis
of cancer if disease is limited to the mediastinum; also sometimes
useful to provide staging of the mediastinal lymph nodes
- 6. thoracoscopy
- a) can use as an initial procedure to be followed by formal
thoracotomy if necessary
- (1) benign Dx in 52%
- (2) complications:
- i) atelectasis 1.2%
- ii) pneumonia 0.8%
- iii) air leak > 7 days 1.6%
- iv) average hospital stay for thorascopy only = 2.4
days
- 7. Surgery
- 8. Screening for lung cancer
- a) initial studies using sputum or chest x-ray did not show
a benefit from screening for lung cancer
- b) a more recent study from Cornell using helical chest CT
scanning in older smokers showed that CT was able to identify
early stage lung cancers, which were presumably more
curable
- c) confirmatory studies involving screening with CT are
underway but preliminary results indicate that in areas of high
histoplasmosis exposure, 25-50% of all patients can have benign
nodules (presumably due to granulomas) which can
radiographically resemble small tumors
- d) pending further studies, most authorities are
withholding endorsement of large scale screening due to
uncertainties regarding the actual beneficial effect of CT
screening on long term outcome, regarding the incidence of
false positive results, and because of the high cost of
screening all at risk persons.
- e) currently, Medicare does not cover screening for lung
cancer in asymptomatic individuals
-
F. Staging System:
-
1. small cell lung cancer: it is controversial whether to use
the TNM system or the limited/extensive system:
- a) limited stage: confined to the hemithorax, mediastinum,
& ipsilateral supraclavicular lymph nodes
- b) extensive stage: any more distant metastasis
- 2. non-small cell lung cancer: use the TNM system for
classification (figure
and table)
- a) definitions:
- (1) Tumor (T)
- TX - primary tumor cannot be assessed
- T0- no evidence of primary tumor
- Tis - carcinoma in situ
- T1 - tumor 3 cm or less, surrounded by lung or
visceral pleura and not in mainstem bronchus
- T2 - tumor > 3 cm OR tumor in mainstem bronchus OR
tumor invading visceral pleura OR tumor with atelectasis
or pneumonia extending to the hilum but not involving the
entire lung
- T3 - tumor invading the chest wall, diaphragm,
mediastinal pleura, or pericardium OR tumor < 2 cm
from the main carina but not involving the main carina OR
tumor with atelectasis or pneumonia involving the entire
lung
- T4 - tumor invading the mediastinum, heart, great
vessels, trachea, esophagus, vertebral body, or main
carina OR tumor with malignant pleural effusion OR tumor
with satellite nodule within the lung
- (2) Lymph nodes (N)
- NX - regional nodes cannot be assessed
- N0 - no regional node involvement
- N1 - ipsilateral peribronchial or hilar nodes
- N2 - ipsilateral mediastinal or subcarinal nodes
- N3 - contralateral mediastinal or hilar nodes OR any
scalene nodes OR any supraclavicular nodes
- (3) Distant metastasis (M)
- MX - distant metastases cannot be assessed
- M0 - no distant metastases
- M1 - distant metastases present
- (b) overall 5-year survival = 13%, partly because the
diagnosis is generally made at an advanced, surgically
incurable stage (figure)
G. Treatment:
- 1. small cell lung cancer -
- a) limited stage (1/3 of cases) - chemotherapy + radiation
therapy:
- (1) typical regimen:
- (a) Etoposide 80-100 mg/m2 IV x 3 days q 21 days for
6 cycles
- (b) Cisplatin 80-100 mg/m2 IV x 1 day
- (c) Plus mediastinal/chest XRT during first 2 cycles
(total ~5-6 weeks, 4500-500 rads)
- (2) Expected survival for limited disease15-25% at 5
years
- (3) Some people recommend prophylactic cranial radiation
for those with complete responses.
- (a) No effect on survival, but decreases brain
relapses.
- b) extensive stage (2/3 of cases) - chemotherapy
- (1) typical regimen:
- (a) Etoposide 80-100 mg/m2 IV x 3 days q 21 days for
6 cycles
- (b) Cisplatin 80-100 mg/m2 IV x 1 day
- (2) expected survival:
- (a) median survival 3 months without treatment
- (b) median survival 9 months with treatment
- (c) even with treatment, essentially no patients
alive at 5 years
- 2. non-small cell lung cancer -
- a) stage I & II - surgery
- (1) PFT assessment of lung function prior to surgery
- (a) FEV1 > 2 L or > 60% predicted
- - predicts sucessful pneumonectomy
- (b) DLCO > 60% predicted
- - predicts sucessful pneumonectomy
- (c) pCO2 > 45
- - predicts increased death rate
- (2) V/Q assessment of lung function
- FEV1 x % perfusion to the lung which will remain
after pneumonectomy > 40% predicted predicts sucessful
pneumonectomy
- (3) exercise assessment of lung function
- (a) maximum oxygen consumption > 20 ml/kg/min
- - predicts low complication rate & death
rate
- (b) maximum oxygen consumption between 10-20
ml/kg/min
- - predicts increased complication rate
- (c) maximum oxygen consumption < 10 ml/kg/min
- - predicts increased death rate
- b) stage IIIA - surgery post-operative radiation therapy
and possibly pre-operative chemotherapy
- c) stage IIIB - high dose radiation (6,000 rads) is
curative in 5-7%; addition of chemotherapy may further increase
survival
- d) stage IV - chemotherapy adds about 8-10 weeks to median
expected survival of 4-5 months
- (1) taxol + cisplatin
- (2) taxol + carboplatin
- (3) cisplatin + navelbine
- (4) cisplatin + gemcitabine
- 3. palliative treatment of lung cancer
- a) external beam radiation therapy
- b) endobronchial brachytherapy (high dose intrabronchial
radiation therapy)
- c) endobronchial stents
- d) endobronchial laser therapy
- e) endobronchial cryotherapy
- f) endobronchial phototherapy
- g) pleurodesis (figure)
H. Palliative Treatment:
- 1. External beam radiation:
- a) If symptoms are mild to moderate and the patient has not
exceeded the maximal radiation dose previously
- b) In patients with more severe respiratory symptoms if
stent placement is not available
- c) In certain situations, radiation may be relatively
contraindicated because of high risk of radiation fibrosis and
brachytherapy, laser, cyrotherapy, or stent may be preferable:
- (1) Co-existant collagen vascular disease
- (2) Recent use of bleomycin, Adriamycin, or
mitomycin
- (3) Pre-existant pulmonary fibrosis
- 2. Brachytherapy:
- a) Technique:
- (1) A bronchoscopic catheter is placed across the
obstructing tumor
- (2) A radioactive iridium source is advanced through the
catheter
- (3) The catheter is withdrawn after Å 15 minutes
- (4) The radiation delivered is high potency but has a
relatively shallow depth of penetration (Å 1 cm)
- b) Indications:
- (1) If symptoms from the obstruction are mild to
moderate
- (2) In general, used only after a full course of
external beam radiation
- (3) If the airway is completely occluded and the patient
can withstand the Å 2 weeks prior to brachytherapy-induced
tumor regression
- (4) In short bronchial segments where a stent will
likely overhang otherwise patent bronchi (eg, right upper
lobe bronchus)
- c) Contraindications:
- (1) Previous maximal brachytherapy to the area
- d) Complications:
- (1) Fibrinous obstruction
- (2) Bleeding
- (3) Fistula formation
- 3. Cryotherapy
- a) Technique:
- (1) Requires a 2.6 mm working channel through a
fiberoptic bronchoscope
- (2) Nitrous oxide cools the cryotherapy probe tip to
-40°C
- (3) One to three 1 minute freeze-thaw cycles
- (4) Can be used for malignant or benign strictures
- b) Advantages:
- (1) Safer than laser
- (2) May be able to cure some carcinoma in situ
- (3) Equipment is less expensive than laser
equipment
- c) Disadvantages:
- (1) Does not provide immediate effect
- (2) May require multiple treatments
- 4. Laser
- a) Indications:
- (1) Airway obstruction
- (2) Bleeding
- (3) Carcinoma in situ
- (a) Can be curative in this situation
- b) Contraindications:
- (1) Extrinsic compression
- (2) Excessively bulky tumor
- (3) Technical difficulty in aiming the laser because of
tumor location
- (4) Upper lobe lesions (can be difficult to
appropriately direct laser due to upward direction of the
lumen)
- (5) Total occlusion of the airway
- c) Complications:
- (1) Hemorrhage
- (2) Pneumothorax
- (3) Respiratory failure
- (4) Fire
- d) Types:
- (1) Carbon dioxide
- (a) Shallow penetration
- (b) Absorbed by water
- (c) Limited hemostasis
- (d) Requires rigid bronchoscope
- (2) Argon
- (a) Mainly used in conjunction with photosensitizing
agents
- (3) Neodymium YAG
- (a) Greatest depth of penetration
- (b) Best hemostasis
- (c) Can be used with rigid or flexible
bronchoscope
-
- 5. Photodynamic therapy
- a) Technique:
- (1) pre-administration of a systemic photosensitizing
agent
- (2) exposure of the airway to very bright white light
using a modified laser
- b) Advantages:
- (1) no risk of ionizing radiation
- (2) can be used pre-operatively to improve bronchial
margins
- (3) very effective for lesions involving the carina
- c) Disadvantages:
- (1) requires laser light source
- (2) risk of severe cutaneous burns if exposed to
sunlight
- (3) shallow depth of penetration
- (4) generally requires extensive forceps
debridement
- (5) typically requires admission to the hospital
- (6) delay from time of administration of
photosensitizing agent to regression of tumor
- 6. Silicone stents:
- 1. Advantage:
- a) Removable
- 2. Disadvantage:
- a) Reduce intralumenal diameter
- b) Retained secretions
- c) Migration
- d) Must be placed with a rigid bronchoscope
- (1) Airway should generally be pre-dilated with rigid
scope prior to stent placement
- 3. Types:
- a) Rüsch Y (Rüsch Inc., Duluth, GA)
- (1) Combined tracheal/bronchial stent with a
"Y"biforcation to permit simultaneous stenting of the
trachea and both main bronchi
- (2) Has the advantage of a silicone body with
stainless steel rings which simulate the tracheal
rings
- (3) In general, "Y" stents are less likely than
straight stents to migrate proximally
- b) Hood bronchial stent (Hood Laboratories, Pembroke,
MA)
- (1) Sialastic stent suitable for areas of tight
stenosis where proximal migration is unlikely
- c) Dumon bronchial stent (Bryan Corp., Woburn, MA)
- (1) Sialastic stent with struts that project into the
bronchial wall to help anchor the stent in place
- d) Montgomery T-tube
- (1) One of the first stents available clinically
- (2) Best suited for upper or mid-tracheal
lesions
- 7. Metal stents:
- 1. Advantages:
- a) Permanent
- b) Generally become covered with new epithelial tissue
consisting of ciliated pseudostratified epithelium within
3-4 months
- c) Can be placed with a flexible bronchoscope
- d) Do not require general anesthesia
- e) Can be done on an outpatient basis
- f) Can be done on ventilated patients in the ICU
- 2. Disadvantages:
- a) Permanent
- b) Can cause compression necrosis if over distended
- c) If they collapse, they can cause airway
obstruction
- d) Tumor can re-grow through the stent with recurrent
obstruction
- 3. Types of metal bronchoscopic stents:
- a) Wallstent (Schneider, Inc., Minneapolis, MN)
- (1) Soft and flexible
- (2) Consist of a woven mesh of stainless steel
- (a) Currently available model consists of 20
braided stainless steel filaments
- (b) A polyurethane covered metal mesh stent is
also available to prevent turmor growth through the
stent mesh
- (3) Self-expanding to a set maximal diameter as an
outer covering sheath is withdrawn
- (4) Do not require a balloon dilation (although this
can be helpful in some circumstances)
- b) Palmaz (Johnson & Johnson Interventional Systems,
Warren, NJ)
- (1) Very stiff
- (2) Metal stent which is loaded over a balloon which
is used to distend the stent to a desired diameter
- (3) It can be irreversibly deformed and crushed with
the force of coughing resulting in airway
obstruction
- c) Gianturco (Cook Inc, Bloomington, IN)
- (1) "Zig-zag" metal stent
- (2) More likely to migrate
- (3) More likely to perforate the bronchus
- (4) Not currently used in the United States
- d) Ultraflex (Boston Scientific, Watertown, MA)
- (1) Soft and flexible
- (2) Made of a single woven nitinol wire (nitinol =
titanium + nickel; nitinol is safe for patients
undergoing MRI)
- (3) Self-expanding to a maximum set diameter
- (4) Requires initial placement of a guidewire via a
flexible bronchoscope
- (5) Does not require balloon dilation
- (6) Has a nylon braid at either end of the stent
which can be grasped by transbronchial forceps for easy
re-positioning of the stent after deployment
- (7) A major advantage over the Wallstent is that as
it expands, it does not shorten, thus making ideal
positioning of the proximal and distal ends of the stent
easier
- 4. Placement of metal stents:
- A. Stent selection
- 1. Of the currently available metal stents, the
Ultraflex and the Wallstent are probably the safest,
easiest to place, and least likely to migrate or
perforate the airway. These are the stents that we use
most commonly at Ohio State
- 2. The decision of whether to use an Ultraflex or a
Wallstent requires experience with both devices. The
Ultraflex is more forgiving for the less experienced
operator but the Wallstent anchors itself into position
better than the Ultraflex. The Wallstent also has greater
radial force than the Ultraflex. At OSU, we require a
minimum of 10 supervised stents for credentialling
purposes.
- B. Pre-stent placement assessment:
- 1. Normal bronchial anatomy:
- a) Trachea = 60 - 90 mm in an adult (intrathoracic
component only)
- (1) Average transverse diameter:
- (a) Women = 15.2 ± 1.4
- (b) Men = 18.2 ± 1.2
- b) Left mainstem bronchus Å 45 mm
- c) Right mainstem bronchus Å 25 mm
- d) Right upper lobe bronchus Å 10 mm
- e) Bronchus intermedius Å 20 mm
- f) Right middle lobe Å 12 mm
- g) Left upper lobe Å 9 mm
- h) Normal bronchial diameters range from 8-12 mm
for main and lobar bronchi
- 2. Radiographic assessment:
- a) Plain chest radiographs
- (1) Occasionally sufficient
- b) Helical CT
- (1) Measure contralateral analagous
bronchus
- (2) Determine if the lesion is calcified (ie,
firm and unlikely to open without balloon
dilation)
- (3) Measure the diameter of the bronchus below
and above the area of obstruction
- (4) Measure the distance between the desired
distal end of the stent and desired proximal end of
the stent
- 3. Bronchoscopic assessment:
- a) Visualize location of lesion relative to patent
bronchi
- b) Determine if lesion is pliable and soft
- c) If bronchoscope can be forced beyond the
lesion, determine the length of the lesion and the
length between patent bronchi by measuring the length
of bronchoscope withdrawn as it is pulled back from
beyond the lesion
- C. Stents are available in multiple sizes:
- 1. 5 - 24 mm diameter by 20 - 90 mm length
- 2. Typical selections:
- a) Bronchus intermedius: 8 x 20 mm
- b) Right mainstem bronchus: 10-12 x 20 mm
- c) Left mainstem bronchus: 10-12 x 20-42 mm
- d) We generally keep 8 x 20, 10 x 20, 12 x 20, 12
x 40, 16 x 60, and 18 x 60 mm stents in stock since
these are the most frequently used stents in our
patient population
- 3. If in doubt about appropriate length, use a
shorter length - a second stent can always be placed
later
- 4. Catheter sizes (in which stents are loaded for
deployment) are measured in French scale
- a) 1 French = 0.33 mm
- b) Transbronchoscopic stent catheters are
generally 7 French (2.3 mm) and thus require at least
a 2.6 mm working channel in the bronchoscope for
passage
- D. Placement issues:
- 1. Positioning
- a) Fluoroscopy must be used
- b) Mark desired distal position of stent with a
radio-opaque marker taped to the chest (eg, a paper
clip)
- 2. If airway is completely or near completely
occluded, use a guidewire to feed the stent through
- a) Flexible tipped guidewires are preferable since
they less likely to perforate the bronchus
- b) Need ³ 200 cm guidewire (generally .035) to
extend the length of the stent and the lesion in front
of it
- c) Guidewires are unnecessary for most lesions,
especially those with a partially patent airway
- 3. If airway if completely or near completely
occluded, consider initial balloon dilation prior to
stent placement
- a) Angioplasty balloons can be fed over the same
guidewire used to feed the stent over
- E. Stent deployment:
- 1. First pass the stent catheter beyond the position
of the obstruction
- 2. The stent is deployed slowly by sliding the the
plastic sheath backward over the metal guide tube
- 3. Initially partially deploy stent (allow about 1/4
to 1/3 of the length of the stent to expand) 4-5 mm
beyond the desired distal final position of the stent and
"snug" the stent up so that the distal end to the stent
coincides with the location in the bronchus previously
marked with the paperclip
- 4. Once the distal end of the stent is appropriately
positioned, slowly deploy the remaining, proximal end of
the stent
- 5. When placing a stent in the trachea
- a) If an endotracheal tube is present, it must be
withdrawn up to a position well above the anticipated
proximal end of the deployed stent so as to not
interfer with the expansion of the stent
- b) Start the deployment in the right mainstem
bronchus and pull the partially deployed distal end of
the stent into the carina so that it is appropriately
positioned in the lower trachea, then fully deploy the
stent
- 6. Inadequate deployment
- a) Results in a stent that is too narrow and too
long (for Wallstents)
- b) Post-deployment balloon dilatation can result
in widening of the stent diameter and shortening of
the stent length to appropriate size
- 5. Complications:
- A. Cough
- 1. Fairly uncommon with metal stents
- 2. Generally only occurs if the stent selected is too
small for the airway lumen
- B. Obstruction of bronchial orifaces
- 1. Often this is an asymptomatic condition although
it can on occasion cause:
- a) Localized wheezing as air passes through the
mesh of the stent
- b) Impaction by secretions wtih dyspnea due to
obstruction of the airway lumen
- c) Pneumonia
- 2. If the stent is so long that it overhangs an
otherwise patent bronchus (eg, a right mainstem stent
extends 6 mm into the trachea and obstructs the left
mainstem bronchus) there are several options for
treatment:
- a) Use transbronchial scissors to make a
longitudinal cut in the stent and allow the stent to
expand open
- (1) In the example, make a 6 mm longitudinal
cut in the proximal left wall of the stent so that
the left mainstem bronchus becomes patent
- (2) This is the easiest option for most
pulmonologists but does require some degree of
extra skill in transbronchial scissor
technique
- b) Laser a hole in the stent
- c) Remove the stent and replace it with a shorter
stent
- (1) Stent removal is difficut if even possible
at all once the stent has become epithelialized;
Ultraflex stents can be removed easir than
Wallstents
- (2) Wallstents require use of rigid
bronchoscope for attempted removal
- C. Secretion retention
- 1. Generally occurs if stent is obstructing a
bronchial oriface
- D. Tumor regrowth through stent mesh with recurrent
obstruction
- 1. Can be avoided by placement of a covered
stent
- E. Proximal migration of stent
- 1. More common with fixed diameter stents (eg,
Palmaz)
- 2. More common when stent placement is followed by
radiation and/or chemotherapy with tumor regression
- F. Obstruction of stent by bent wires
- 1. Can occur due to suction catheters, etc
- 2. Wires can be bent back into appropriate position
by transbronchial biopsy forceps
- G. Erosion into nearby blood vessels
- 1. More common with rigid metal stents (eg, Strecker
or Palmaz)
- 2. Stents should be used cautiously if at all when
airway obstruction is due to compression by a vessel
- 8. Personal preferences for malignant airway obstruction:
- a) Use external beam radiation for initial management of
mild-moderate symptoms of malignant obstruction
- b) Use brachytherapy, cryotherapy, or laser for patients
failing external beam radiation or for patients in whom
external beam radiation is contraindicated
- c) Use stents for patients with acute, severe symptoms of
malignant obstruction and for patients who have already had
brachytherapy and/or cryotherapy
I. The Solitary Pulmonary Nodule (figure):
- 1. solitary pulmonary nodules
- a) smallest lesion detectable with CXR is 2-3 mm
- b) 40-50% malignant
- (1) 10-30% of these are metastatic
- c) 40-50% granulomas
- d) benign calcium patterns (often better visualized with CT
than plain chest x-rays):
- (1) laminated
- (2) diffuse
- (3) central
- (4) note: eccentric or speckled can be seen with
malignant or benign
- (a) 13% of bronchogenic cancers have CT evidence of
calcification (eccentric or speckled)
- e) malignant nodules have a doubling time of 21 - 400 days
(figure)
- (1) a lesion which is unchanged in size for ³ 2 years is
usually benign
- f) indicators of malignancy:
- (1) size
- (a) < 2 cm = 20% malignant
- (b) > 3 cm = 80% malignant
- (2) age
- (a) < 50 years old = 33% malignant
- (b) > 50 years old = 65% malignant
- (3) smoking
- (4) history of cancer
- g) PET (positron emission tomography) can be useful to
distinguish benign from malignant pulmonary nodules when biopsy
is medically risky (figure):
- (1) sensitivity = 96%
- (2) specificity = 88%
I. References:
- 1. Am. J. Respir. Crit. Care Med., Volume 156, Number 1, July
1997, 320-332
- 2. Chest, Volume 111, Number 6, June 1997, 1710-1717
- 3. Chest, Volume 112, Number 4S, October 1997, 251-258
- 4. Chest, Volume 115, Number 1, January 1999, 233-235
- 5. N. Engl. J. Med. 1994; 330:159-164
- 6. Soc. Sci. Med. 1991; 32:1151-1159
- 7. JAMA 1994; 271:1752-1759
- 8. http://cancernet.nci.nih.gov
- 9. New England Journal of Medicine 343(22): 1627-1633,
2000.
-
-
- last updated May 2, 2001
-
- TOP OF
PAGE
-
- Return to the For
Residents Page