(Chest. 2001;119:788-794.)
© 2001
American College of Chest Physicians
Hemoptysis in Patients With Renal Insufficiency*
The Role of Flexible Bronchoscopy
Nicholas Kallay, MD;
Donnie P. Dunagan, MD, FCCP;
Norman Adair, MD, FCCP;
Robert Chin, MD, FCCP and
Edward F. Haponik, MD, FCCP
*
From the Department of Internal Medicine (Drs. Kallay, Dunagan, Adair, and Chin), Section on Pulmonary and Critical Care, Wake Forest University Baptist Medical Center, Winston-Salem, NC; and the Division of Pulmonary and Critical Care Medicine (Dr. Haponik), Johns Hopkins University School of Medicine, Baltimore, MD.
Correspondence to: Nicholas Kallay, MD, Pulmonary Fellow, Section on Pulmonary and Critical Care Medicine, Wake Forest University Baptist Medical Center, Medical Center Blvd, Winston-Salem, NC 27157-1054; e-mail: nkallay{at}wfubmc.edu
 |
Abstract
|
|---|
Study objectives: To assess the indications, yield, and
therapeutic impact of flexible bronchoscopy (FB) in patients with
hemoptysis and renal insufficiency.
Design:
Retrospective cohort analysis.
Setting: Tertiary-care
university hospital.
Patients: Thirty-four patients
over a 7.5-year period who underwent FB to evaluate hemoptysis in the
setting of renal insufficiency (ie, serum creatinine
level, > 1.5 mg/dL).
Measurements and results: The
etiology of hemoptysis was undetermined in 41% of cases. Defined
causes of bleeding included infections (29%), pulmonary renal
syndromes (15%), airway injury (9%), and pulmonary embolism (6%). No
specific bleeding site was identified, but FB lateralized hemorrhaging
to one lung in 24% of patients. FB results influenced therapy in 29%
of patients overall and in 8% of patients without respiratory tract
infection. The hospital survival rate was 47% and did not differ based
on the presence or absence (presence vs absence) of the following
variables: a defined etiology for hemoptysis (45% vs 50%);
lateralized bleeding (38% vs 50%); or management alterations prompted
by other FB findings (50% vs 46%). Factors associated with survival
included the onset of bleeding prior to hospital admission (80% vs
33%; p = 0.02), the absence of respiratory failure requiring
mechanical ventilation at the time of FB (90% vs 29%; p = 0.002),
and lack of prohemorrhagic factors (other than uremia) such as
disseminated intravascular coagulation, recent treatment with
warfarin, heparin, or antiplatelet agents (78% vs 33%; p = 0.05).
During the 6 months following hospital discharge, hemoptysis recurred
in 14% of patients, and 5 patients died, for an overall mortality rate
of 62%.
Conclusions: These data suggest that FB in
hospitalized patients with hemoptysis and renal insufficiency, and
without radiographic findings suggesting neoplastic disease, has a low
yield and limited impact. Whether FB influences outcome in selected
patients in this setting requires prospective
investigation.
Key Words: bronchoscopy • hemoptysis • human • kidney failure • outcome assessment
|
Introduction
|
|---|
In
patients with hemoptysis, flexible bronchoscopy (FB) is commonly used
to help determine a specific etiology, to localize bleeding, and to
guide therapeutic decisions.1
2
3
Hemoptysis has a myriad
of potential causes, and its severity may range from
inconsequential bleeding to an acute, life-threatening
event.4
5
In patients with renal failure, important and
potentially treatable considerations include pulmonary renal syndromes,
most commonly, Wegener’s granulomatosis, Goodpasture’s syndrome, and
systemic lupus erythematosus. Moreover, such patients may have
increased predispositions toward
infection,
bleeding diatheses, and iatrogenic factors, as well as the spectrum of
conditions seen in persons without renal failure. Although FB is
performed frequently in the evaluation of hemoptysis, its role in
evaluating patients with coexisting renal insufficiency has not
been defined. We reviewed our experience with FB in this clinical
setting to better characterize this patient population, to
clarify the role of FB and its clinical impact, and to appraise whether
clinical characteristics may be associated with improved outcome.
|
Materials and Methods
|
|---|
Patient Selection
The study was designed as a retrospective cohort analysis. Using
a database of hospital billing records, all patients discharged from
Wake Forest University Baptist Medical Center between January 1, 1990,
and June 30, 1998, who had been evaluated with FB and had been given a
diagnosis of renal insufficiency (defined as a serum creatinine level
> 1.5 mg/dL, the upper limit of the normal range in our hospital
laboratory) and who experienced hemoptysis at any time during
hospitalization were identified. Medical records for these patients
were examined to identify those in whom FB was performed to evaluate
hemoptysis with concurrent renal insufficiency.
Data Collection
All data regarding the patients were obtained directly from
hospital databases, the patient’s medical record, and FB reports, and
included patient demographics, the primary reason for hospital
admission, and laboratory studies at presentation or within 24 h
of hemoptysis onset (ie, WBC count, hemoglobin count,
platelet count, prothrombin time/international normalized ratio
[INR], partial thromboplastin time, BUN level, serum creatinine
level, PaO2, oxygen
saturation as measured by pulse oximetry
[SpO2], and chest
roentgenogram). Relevant clinical features were noted, including the
timing of the onset of hemoptysis relative to hospital admission, the
volume of bleeding, the etiology of renal failure and hemoptysis, the
need for mechanical ventilation at the time of FB, and the use of drugs
or existence of conditions that might predispose the patient to
hemorrhage (in addition to uremia). FB findings, including airway
examinations and findings from BAL and protected specimen brush (PSB)
studies, were recorded. Mortality at the time of hospital discharge was
noted. In addition, follow-up data regarding recurrence(s) of bleeding,
hospitalization, and survival for the 6 months following discharge were
obtained.
FB
All patients were seen in consultation by members of the
pulmonary faculty. Prospective criteria for patient selection for FB
were not defined, but general indications favoring the performance of
FB in patients with hemoptysis at our center include increasing volume
and duration of bleeding, as well as the presence and pattern of
abnormal chest radiograph findings. Contraindications to FB included
hemodynamic instability, oxygenation status that would be significantly
compromised by FB, acute myocardial ischemia, and life-threatening
cardiac arrhythmia.
Airway examination was performed in all patients to determine the
location and etiology of the hemorrhage. If the specific bleeding site
was not reported, information regarding a more general origin of
bleeding (ie, localization to an individual lobe or
right/left lateralization) was determined from the FB report. Specimen
sampling during the procedure was individualized for each case and
included BAL and PSB. In general, BAL and PSB specimens were submitted
for quantitative bacterial culture, fungal smear and culture, viral
culture, mycobacterial studies, direct fluorescent antibody test for
Pneumocystis carinii, and cytology testing. Techniques for
microbiological specimen processing and interpretation have been
reported previously.6
The results of bacterial cultures
were considered to be positive if an organism was isolated in numbers
10,000 cfu/mL from BAL specimens or 1,000 cfu/mL from PSB
specimens.
Outcome Measures
The impact of FB on patient management was evaluated by a review
of all progress notes and physician orders following the procedure.
Treatment changes were recorded if it was clear that they had been made
based on FB results. Potential therapeutic modifications included, but
were not limited to, changes in antimicrobial therapy, the addition or
discontinuation of immunosuppressive medications, interventional
angiography, and/or surgical treatment.
In-hospital mortality data were obtained for all patients.
Relationships were sought between survival and a number of variables,
including the presence or absence of hemoptysis at the time of hospital
admission, whether FB was performed before or after endotracheal
intubation (if needed), the presence or absence of exposure to drugs or
conditions predisposing the patient to hemorrhage (other than uremia),
and therapeutic changes based on FB results. Further outcome data were
sought for all patients surviving to hospital discharge to determine
the reported recurrence of hemoptysis, readmission rates (and
indications), subsequently diagnosed etiologies of hemoptysis, and
mortality during the 6 months following discharge.
Data Analysis
Data were analyzed using a standard statistical package (SAS;
SAS Institute; Cary, NC). Descriptive statistics such as means, SDs,
and frequencies were generated for the variables of interest. Fisher’s
Exact Test was used to test for the significance of dichotomous
variables. Wilcoxon rank sum test was used to assess the association
between continuous variables and the outcomes of interest. A p value
0.05 was considered to indicate statistical significance.
|
Results
|
|---|
Patient Selection, Demographics, and Laboratory Data
During the period of review, we identified 13,145 patients with
renal failure, 365 (2.8%) of whom also experienced hemoptysis. One
hundred seventeen patients (0.9%) were evaluated with FB and
experienced hemoptysis at some time during their hospitalization. Of
these patients, 115 sufficiently complete patient medical records could
be found and were reviewed to extract 34 patients (29%) in whom these
three characteristics were present simultaneously.
Demographic data, the primary reasons for hospital admission, and the
overall survival rate at discharge are summarized in Table 1
. The most common reasons for hospitalization included hemoptysis, renal
failure, and respiratory failure (24 patients, 71% of total), but
there was considerable variability in other less common precipitants.
Eight patients were admitted with hemoptysis as their presenting
complaint; however, two additional patients later admitted that they
had experienced hemoptysis before their hospitalization. There was a
wide range in the duration (1 day to 2 months) and volume (from trace
amounts to 200 mL/d) of hemoptysis. None of the patients were
described as having "massive hemoptysis" or volumes exceeding 200
mL/d. The all-cause mortality rate was high in this selected
population, as 16 patients (47%) survived and 18 patients (53%) died.
The precise cause of death was often unclear, but chart review
suggested that hemoptysis was the direct cause of death in only one
patient. This individual had bronchoscopically diagnosed aspergillus
pneumonia, underwent a previously scheduled tracheostomy following FB,
and died the following day from massive airway hemorrhage that was
thought to be related to his tracheostomy site.
Laboratory and radiographic data obtained at hospital admission or
within 24 h of hemoptysis onset are presented in Table 2
. In five patients (15%), chest radiographs demonstrated unilateral
findings. Of the remaining radiographs, 18 (82%) demonstrated
bilateral opacities in the lung fields, and 1 (3%) was interpreted as
normal. Only one patient had a radiographic suggestion of neoplastic
disease, a peripheral nodule.
Four clinical measures correlated with outcome. Increased
SpO2 (95% vs 82%; p = 0.036) and
decreased oxygen requirements (fraction of inspired oxygen
[FIO2] 0.34 vs 0.54; p = 0.003)
at the time of hemoptysis were associated with survival. Serum
creatinine levels were higher in survivors than in nonsurvivors (6.9 vs
3.0 mg/dL; p = 0.001); however, all four patients admitted to the
hospital with preexisting end-stage renal disease (mean serum
creatinine level, 8.0 mg/dL), as well as a single patient admitted with
a creatinine level of 23 mg/dL, survived to discharge, possibly skewing
these data. Measurements of mean prothrombin time, INR, activated
partial thromboplastin time, and platelet count did not discriminate
between patients who did or did not survive. However, the 73% of our
patients having one or more prohemorrhagic factors (ie, the
presence of exposure to warfarin, heparin, or antiplatelet agent during
the preceding 24 h and/or the presence of disseminated
intravascular coagulation or other coagulopathy) proved to have a lower
survival (33% vs 78%; p = 0.05).
Etiologies of Renal Failure and Hemoptysis
The etiologies of renal failure are described in Figure 1
. In only five patients (15%) was renal failure thought to be due to a
pulmonary/renal syndrome, including Wegener’s granulomatosis (three
patients), systemic lupus erythematosus (one patient), and idiopathic
pulmonary renal syndrome (one patient, in whom serum antineutrophilic
cytoplasmic antibody showed a positive perinuclear pattern and renal
biopsy specimen demonstrated immune complex glomerulonephritis). Other
causes of renal failure were multiple myeloma, hypertensive
nephrosclerosis, and otherwise unspecified glomerulonephritis.
As summarized in Figure 2
, the precise cause of hemoptysis frequently remained unclear (41% of
patients) at the time of patient death or hospital discharge.
Infectious etiologies accounted for 29% of instances, a pulmonary
renal syndrome was thought to be the only etiology in 15%, airway
injury in 9% (two patients with suction catheter trauma and one
patient with pulmonary contusion), and pulmonary embolism in 6%.
FB Results
FB was performed most often at least 48 h after the onset of
hemoptysis, when deemed appropriate and safe by the consulting and
attending physicians. Although all patients had hemoptysis, other
indications for FB were also present in 28 patients (82%), 24 of whom
had clinically suspected respiratory tract infections. The relative
weight assigned to these factors in the decision for FB could not be
determined with certainty, but chart review confirmed that hemoptysis
figured prominently in the decision to proceed with FB in all cases. In
particular, increased bleeding was not observed except in the
previously mentioned instance following scheduled tracheostomy. FB
findings are summarized in Table 3
. Neither a specific endobronchial etiology nor a focal site of bleeding
was identified during FB in any patient. Blood was visualized in 88%
of patients, and the most common finding was bilateral hemorrhage
(50%). Progressively bloody lavage fluid suggesting diffuse alveolar
hemorrhage was observed in a single patient. Two of these patients
underwent transbronchial forceps lung biopsy, the results of both of
which were nondiagnostic. FB cultures were positive in 15
patients (63%) and revealed the following broad range of
microorganisms: Staphylococcus epidermidis (n = 1);
Staphylococcus aureus (n = 3); Streptococcus
pneumoniae (n = 1); Stenotrophomonas maltophilia
(n = 2); Haemophilus parainfluenzae (n = 1);
Pseudomonas aeruginosa (n = 1); Mycobacterium
avium (n = 1); Candida species (n = 5); Aspergillus (n = 1);
Pneumocystis carinii (n = 2); herpes simplex virus
(n = 1); cytomegalovirus (n = 1); and adenovirus (n = 1). The
results of cytology testing in one instance were thought to be
suspicious for small cell carcinoma (in the absence of endobronchial
abnormality), but the patient died before a definite diagnosis could be
made, and an autopsy was not obtained.
Relationship of FB to Outcomes
Overall, FB did not appear to have a substantial impact on
outcome. Identification of the etiology of hemoptysis was not
associated with improved survival (45% in known vs 50% in unknown
etiologies; the p value was not significant [NS]), nor was
lateralization of bleeding (38% when lateralized vs 50% when not
lateralized; NS). The 10 patients (29%) identified as having
hemoptysis prior to hospital admission were more likely to live than
those in whom hemoptysis began after admission (80% vs 33%;
p = 0.02).
In 10 patients (29%), therapy was changed on the basis of
bronchoscopic findings. These changes included eight cases of
modifications to antibiotic regimens, one case of angiography/bronchial
artery embolization, and one case of minimizing endotracheal tube
suctioning. Thus, FB influenced therapy in only 2 of 24 noninfected
patients (8.3%). When patient therapy was changed by FB results, the
survival to discharge rate (50%) was similar to that in patients who
did not have therapy changed (46%). When FB was performed in 24
patients after the initiation of mechanical ventilation, survival was
lower than in the 10 patients who received FB prior to a need for
mechanical ventilation (29% vs 90%; p = 0.002) (Fig 3
). There were no recorded complications resulting from FB.
Postdischarge Follow-up
Of the 16 patients who had been discharged alive from the
hospital, 14 (87.5%) had detailed follow-up records available that
spanned at least 6 months. Ten patients required a total of 15
additional hospitalizations in the 6 months following their initial
discharge. Three admissions (20%) were caused by pneumonia, three
admissions (20%) were caused by pulmonary edema, two admissions (13%)
were caused by complications relating to hemodialysis access, and one
admission (6.7%) each was caused by chest pain, vertebral
osteomyelitis, spontaneous pneumothorax, bilateral vocal cord
paralysis, severe bronchitis, following cardiopulmonary arrest while
receiving hemodialysis, and with hemoptysis not otherwise specified.
The recurrence of small amounts of hemoptysis was described in 2 of 14
patients (14%) in the 6 months after their initial hospitalization. In
one patient, bleeding was attributed to pneumonia, while the other
patient received a diagnosis of idiopathic pulmonary hemosiderosis
following an otherwise nondiagnostic open lung biopsy.
Survival status 6 months after hospital discharge could be determined
in all 16 discharged patients. Five patients (31%) died in the 6
months following their hospital discharge, for an overall mortality of
rate of 62% (21 of 34 patients). Deaths were attributed to pneumonia
(three patients, one of whom had presented with small amounts of
hemoptysis), sepsis complicating a hospitalization for spontaneous
pneumothorax, and cardiopulmonary arrest of unclear etiology while
receiving hemodialysis.
|
Discussion
|
|---|
Hemoptysis and renal failure are common clinical conditions with
potentially important morbidity and mortality, but reports about their
simultaneous occurrence are generally limited to those in patients with
pulmonary renal syndromes. This report is the first to
review the use of FB in patients selected on the basis of the presence
of concurrent hemoptysis and renal failure. While the etiology of
hemoptysis often is established, earlier reports have described
hemoptysis as cryptogenic in 3 to 22% of patients.7
The
frequency of idiopathic hemoptysis (41% after hospitalization, 35%
after follow-up) in our patients is high and may be due, at least in
part, to a greater incidence of comorbidities, including concomitant
renal insufficiency. In a large proportion of our patients (73%),
coexisting factors (in addition to uremia) or recent treatment with
medications that might increase the tendency to bleed were identified.
Importantly, all but one of these patients had no radiologic findings
suggesting neoplastic disease.
The coexistence of hemoptysis and renal insufficiency often raises
concerns about pulmonary renal syndromes, but these conditions were
found in only 15% of our patients. This group of diseases
includes Wegener’s granulomatosis, Goodpasture’s syndrome, and
systemic lupus erythematosus, as well as less common entities such as
Behçet’s syndrome and Henoch-Schonlein purpura, all of which may
have a component of pulmonary capillaritis that may result in diffuse
alveolar hemorrhage.8
9
10
The FB finding of increasingly
bloody return after the instillation of successive aliquots of saline
solution lavage has been associated with alveolar hemorrhage, and a
pathologic diagnosis of pulmonary capillaritis can be made with
transbronchial forceps biopsy.11
Two of our patients
underwent transbronchial forceps biopsy, which was not diagnostic in
either patient. The only patient with progressively bloody lavage fluid
return eventually proved to have an idiopathic etiology of bleeding
despite extensive evaluation. Importantly, it is likely that the low
frequency of pulmonary renal syndromes in this review reflects the
availability of serologic markers together with institutional
philosophies favoring renal biopsy and/or lung biopsy (open or
thoracoscopic), rather than bronchoscopic evaluation, when these
conditions are most strongly considered.
A major goal in the evaluation of hemoptysis is to localize the site of
bleeding.12
Such information may help to guide decisions
regarding surgery, interventional angiography, and effective airway
control. In our patients, FB did not identify any discrete source of
bleeding and demonstrated lateralizing findings in only 24% of cases,
without an apparent impact on survival. Most of our patients underwent
FB > 48 h after the onset of hemoptysis, potentially influencing the
diagnostic yield. In the report of Gong and Salvatierra13
on 129 patients, early FB (during active hemoptysis or within 48 h
of cessation) was associated with an increased likelihood of
visualizing active bleeding or its site, but clinical outcomes were
similar compared to those patients in whom FB was delayed. In addition,
although FB contributed to a modification of therapy in nearly a third
of our patients, survival did not appear to be influenced by these
changes.
Thus, knowledge of the etiology of hemoptysis (when determined),
localization of the bleeding site, and FB-directed therapeutic changes
in this patient series had no apparent impact on patient survival. This
observation suggests that in patients with renal failure the need for
generalized respiratory supportive measures and the clinical course of
associated comorbidities, rather than the hemoptysis itself, may have
the greatest impact on short-term patient survival. Although none of
our patients with renal insufficiency had hemoptysis that would
generally be described as massive14
15
(when volumes were
reported), their overall mortality rate was high (53%). This
observation suggests that associated comorbidities accounted for many
of the deaths in our group of patients.
Five other clinical markers were associated with survival in these
selected patients. The onset of hemoptysis prior to hospital admission
was associated with an increased likelihood of survival, perhaps
reflecting a relative lack of coexisting illnesses in these patients
compared to those hospitalized patients in whom hemoptysis was
superimposed on other critical problems (including acute renal
failure). Similarly, the ominous outcome for patients with a lower
SpO2 level, a higher
FIO2 requirement, or the need for
mechanical ventilation prior to FB is consistent with the underlying
severity of illness rather than solely the hemoptysis that they
experienced. The absence of any prohemorrhagic factors (ie,
exposure to warfarin, heparin, or antiplatelet agent during the
preceding 24 h, and/or the presence of disseminated intravascular
coagulation or other coagulopathy at the time of FB) was associated
with survival. The individual elements of this category might be
expected to vary in their prognostic implications, but the small size
of our series limits this analysis.
The 6-month follow-up of patients who survived their initial
hospitalizations revealed a modest 14% recurrence rate of small-volume
hemoptysis and a more substantial 31% mortality. While one of the five
deaths was associated with a recurrence of small-volume hemoptysis in
the setting of pneumonia, the other four deaths had no clear link to
the hemoptysis experienced during the initial hospitalization.
Follow-up yielded only one additional etiology for the patient’s
initial hemoptysis. It is noteworthy that no diagnoses of neoplastic
disease were made during this period.
Several limitations to this report must be acknowledged and include
those inherent to retrospective studies. There were no a
priori criteria used to identify candidates for bronchoscopy, and
complete data sets were not always available for all patients. While
uniform prospective criteria for FB were not defined, bleeding had to
be sufficient in volume and/or duration to merit FB. Patients with
brief, self-limited episodes and low volumes of hemoptysis generally
would not receive FB. Despite selection criteria that might have
favored FB, its yield and apparent impact were low. Our patients tended
to have multiple comorbidities, prolonged hospitalizations, and, in 28
cases, more than one potential indication for FB. We can only speculate
as to whether these patients would have undergone FB in the absence of
hemoptysis. The inclusion of patients with multiple indications for FB
might allow greater opportunities for the procedure to demonstrate its
worth. The prognostic impact of the volume of hemoptysis could not be
determined because reliable quantitation of bleeding was inconsistently
recorded. In addition, we did not appraise the impact of nonspecific or
negative results of FB. Such findings have the potential to
influence management,16
but these effects could not be
measured readily.
|
Comments
|
|---|
Despite these limitations, the current findings suggest that in
most hospitalized patients with hemoptysis and renal failure, a
specific endobronchial etiology of bleeding is often (41%) undefined,
hospital mortality is high (53%), and short-term survival is not
altered substantially by bronchoscopic information. A known etiology of
hemoptysis, the presence of lateralizing bronchoscopic findings, and
alteration of therapy by FB did not appear to alter outcome. Factors
associated with survival included the onset of hemoptysis prior to
hospital admission, the performance of FB prior to mechanical
ventilation, a preserved level of oxygen saturation with low
FIO2 requirements, a high serum
creatinine level, and the absence of recent exposure to prohemorrhagic
factors (other than uremia) at the time of hemoptysis onset. Follow-up
over 6 months revealed that although additional etiologies for
hemoptysis seldom were established after hospital discharge (7%) and
the frequency of recurrent hemoptysis was low (14%), the mortality
rate was nevertheless substantial (31%). Deaths during and shortly
after hospitalization often appear to be due to reasons other than
hemoptysis, likely representing other comorbidities. These data suggest
that FB in hospitalized patients with concurrent renal failure and
hemoptysis, without radiographic findings suggesting neoplastic
disease, has a low yield and a limited impact on outcome. Whether FB in
selected patients with renal failure is beneficial requires prospective
investigation.
|
Footnotes
|
|---|
Abbreviations: FB = flexible bronchoscopy;
FIO2 = fraction of inspired oxygen;
INR = international normalized ratio; NS = not significant;
PSB = protected-specimen brush;
SpO2 = oxygen saturation as measured by pulse
oximetry
Received for publication July 27, 1999.
Accepted for publication July 17, 2000.
|
References
|
|---|
-
Imgrund, SP, Goldberg, SK, Walkenstein, MD, et al (1985) Clinical diagnosis of massive hemoptysis using the fiberoptic bronchoscope. Crit Care Med 13,438-443[ISI][Medline]
-
Smiddy, JF, Elliott, RC (1973) The evaluation of hemoptysis with fiberoptic bronchoscopy. Chest 64,158-162[Abstract/Free Full Text]
-
Zavala, DC (1975) Diagnostic fiberoptic bronchoscopy. Chest 68,12-19[Abstract/Free Full Text]
-
Santiago, S, Tobias, J, Williams, AJ (1991) A reappraisal of the causes of hemoptysis. Arch Intern Med 151,2449-2451[Abstract]
-
Haponik, EF (1997) Hemoptysis. Kelly, WN eds. Textbook of internal medicine 3rd ed. ,1918-1923 Lippincott-Raven Philadelphia, PA.
-
Baker, AM, Bowton, DL, Haponik, EF (1995) Decision making in nosocomial pneumonia: an analytical approach to the interpretation of quantitative bronchoscopic cultures. Chest 107,85-95[Abstract/Free Full Text]
-
Hirshberg, B, Biran, I, Glazer, M, et al (1997) Hemoptysis: etiology, evaluation, and outcome in a tertiary referral hospital. Chest 112,440-444[Abstract/Free Full Text]
-
Green, RJ, Ruoss, SJ, Kraft, SA, et al (1996) Pulmonary capillaritis and alveolar hemorrhage: update on diagnosis and management. Chest 110,1305-1316[Abstract/Free Full Text]
-
Kelly, PT, Haponik, EF (1994) Goodpasture syndrome: molecular and clinical advances. Medicine 73,171-185[Medline]
-
Zamora, MR, Warner, ML, Tuder, R, et al (1997) Diffuse alveolar hemorrhage and systemic lupus erythematosus: clinical presentation, histology, survival, and outcome. Medicine 76,192-202[CrossRef][Medline]
-
Green, RJ, Ruoss, SJ, Kraft, SA, et al (1996) Pulmonary capillaritis and alveolar hemorrhage: update on diagnosis and management. Chest 110,1305-1316
-
Cahill, BC, Ingbar, DH (1994) Massive hemoptysis: assessment and management. Clin Chest Med 15,147-168[ISI][Medline]
-
Gong, H, Salvatierra, C (1981) Clinical efficacy of early and delayed fiberoptic bronchoscopy in patients with hemoptysis. Am Rev Respir Dis 124,221-225[ISI][Medline]
-
Knott-Craig, CJ, Oostuizen, JG, Rossouw, G, et al (1993) Management and prognosis of massive hemoptysis: recent experience with 120 patients. J Thorac Cardiovasc Surg 105,394-397[Abstract]
-
Israel, RH, Poe, RH (1987) Hemoptysis. Clin Chest Med 8,197-205[ISI][Medline]
-
Haponik, EF, Chin, R (1990) Hemoptysis: clinicians’ perspectives. Chest 97,469-475[Abstract/Free Full Text]
TOP
Abstract
Introduction
