atient characteristics associated with bronchial artery ...pennir.com/content/bronchial artery embo...

PATIENT CHARACTERISTICS ASSOCIATED WITH BRONCHIAL ARTERY

EMBOLIZATION OUTCOMES

An Evidence Review from Penn Medicine’s Center for Evidence-based Practice

June 2015

Project director: ..................... Kendal Williams, MD, MPH (CEP) Lead analyst: ....................... Matthew D. Mitchell, PhD (CEP) Internal review:...................... Nikhil Mull, MD (CEP)

Keywords: hemoptysis, embolization, tuberculosis, aspergillosis, cystic fibrosis

EVIDENCE SUMMARY Published clinical studies of bronchial artery embolization (BAE) for treatment of massive or life-threatening hemoptysis

used differing thresholds for determining which patients needed this treatment. There were no additional patient selection

criteria in these studies that could be used in developing evidence-based guidelines for use of this procedure.

Most patient characteristics including age, sex, underlying disease, and comorbidity do not appear to have an effect on the

rate of recurrent hemoptysis after BAE. The evidence for this conclusion is weak; all of the evidence came from retro-

spective cohort studies conducted in Asia or South Africa. This evidence may be less applicable to patients and practice in

the United States.

There is very weak evidence that patients with aspergillosis are at greater risk of recurrent hemoptysis after BAE than

patients whose hemoptysis is caused by other diseases. There is no evidence on recurrence risk specific to patients with

cystic fibrosis or with interstitial lung diseases other than tuberculosis. There is also little or no evidence relating to possible

differences between groups in procedure success rates, rate of complications, or other relevant outcomes.

Patients whose hemoptysis is of highest volume (at least 400 ml per 24 hours or 200 ml per event) are at greater risk of

recurrent hemoptysis after BAE; though evidence for this conclusion is inconsistent and very weak. Patients with a history

of previous hemoptysis are also at greater risk of recurrent hemoptysis; the evidence for this conclusion is weak. A history

of previous BAE was not associated with increased risk of recurrent hemoptysis. There is a possible increase in recurrence

risk for patients with a bronchio-pulmonary shunt or other abnormal vascular anatomy.

© Copyright 2015 by the Trustees of the University of Pennsylvania. All rights reserved. No part of this publication

may be reproduced without permission in writing from the Trustees of the University of Pennsylvania. R300

CEP Evidence Review: Bronchial artery embolization 2

Table of Contents

Introduction ................................................................................................................................................................................................. 3

Previous CEP reports .............................................................................................................................................................................. 3 Methods....................................................................................................................................................................................................... 4

Protocol for Systematic Review .............................................................................................................................................................. 4

Literature Search ..................................................................................................................................................................................... 5 Table 1. Medline search ..................................................................................................................................................................... 5

Table 2. EMBASE search .................................................................................................................................................................. 6 Table 3. Cochrane Library search ...................................................................................................................................................... 6

Results ......................................................................................................................................................................................................... 7

Guidelines ............................................................................................................................................................................................... 7 Reviews ................................................................................................................................................................................................... 7

Primary literature .................................................................................................................................................................................... 7 Risk of recurrent hemoptysis .............................................................................................................................................................. 7 Table 4. Primary studies: recurrence of hemoptysis ........................................................................................................................ 8

Other outcomes ................................................................................................................................................................................. 10 Conclusion ................................................................................................................................................................................................ 11

Table 5. Evidence summary and GRADE analysis ......................................................................................................................... 11

References ................................................................................................................................................................................................. 13

Appendix. GRADE criteria for rating a body of evidence on an intervention ........................................................................................ 14


Introduction

There is disagreement among clinicians over how much hemoptysis is necessary for it to warrant immediate action. Concerns with

overuse of bronchial artery embolization (BAE) and its risks to patients cause some to argue for higher thresholds and selective use of

BAE while concerns with the consequences of continued bleeding cause others to argue for lower thresholds and broader use of the

procedure. Published clinical guidelines vary in the threshold defining “massive” or “life-threatening” hemoptysis, and they do not

cite specific evidence to support their thresholds. The purpose of this review is to search for and analyze clinical studies which might

be used to develop evidence-based guidelines for use of bronchial artery embolization and other interventions to treat massive or life-

threatening hemoptysis.

Previous CEP reports

The Center for Evidence-based Practice has published an Annotated Bibliography on algorithms for management of patients with

massive hemoptysis (1) and an Evidence Advisory on guidelines for management of patients with massive hemoptysis (2). Neither

report found evidence-based guidance for deciding which patients should or should not be treated with bronchial artery embolization.


Methods

CENTER FOR EVIDENCE-BASED PRACTICE

PROTOCOL FOR SYSTEMATIC REVIEW SPECIFIC AIM:

Identify patient groups for whom bronchial artery embolization is especially effective, especially ineffective, or for whom potential risk outweighs potential benefit.

METHODS:

Inclusion and exclusion criteria:

Participants: Patients with hemoptysis, particularly massive or life-threatening hemoptysis. Report subgroup of patients with cystic fibrosis if data permits.

Interventions: Bronchial artery embolization (BAE).

Comparisons: Subgroups of patients with hemoptysis such as those with particular demographics or comorbidities, those with differing causes of hemoptysis, and those with differing degrees of hemoptysis.

Outcomes: Mortality, paralysis, other procedure-related adverse events, rebleeding, need for second procedure, length of stay, length of ICU stay.

Data collection

Databases: Cochrane databases, Medline, EMBASE.

Study design: All clinical studies, though RCTs will be given priority.

Study quality assessment: Randomized trials (if any) assessed using modified Jadad scale.

Data synthesis (calculation of relative risks and confidence intervals, meta-analyses, exploration of heterogeneity): Random-effects meta-analysis following Cochrane methods if quantity and homogeneity of data permit, otherwise qualitative analysis.

Assessment of quality of evidence base: GRADE (Appendix).


Literature Search

Searches were completed in May and June 2015 and were not restricted by date. Medline searches made use of diagnosis

subcategories of the VTE indexing terms. Guideline sources were searched in our previous report on guidelines for management of

massive hemoptysis, where we found that none of the thresholds in published guidelines were based on specific evidence. Reference

lists of published review articles were also checked: this resulted in four additional articles retrieved, but those articles did not report

any risk factor information.

Table 1. Medline search

Search Syntax Hits Retrieved Included

1 Embolization, Therapeutic/ 24,817 — —

2 (bronchial or hemoptysis or haemoptysis).mp. 100,979 — —

3 1 and 2 830 — —

4 (bronchial adj3 (emboliz* or embolis*)).mp. 513 — —

5 3 or 4 1,000 — —

6 (guideline or guidance).mp. or exp Guideline/ or exp Practice Guideline/ 139,917 — —

7 5 and 6 11 3 0

8 limit 5 to (meta analysis or systematic reviews) 5 — —

exclude 3 references included in guideline results 2 1 0

9 limit 5 to (clinical trial, all or clinical trial or controlled clinical trial or

pragmatic clinical trial or randomized controlled trial)

15 — —

10 exp Embolization, Therapeutic/ae, ct, mo, sn [Adverse Effects, Contraindications,

Mortality, Statistics & Numerical Data]

4,409 — —

11 2 and 10 112 — —

12 9 or 11 124 — —

exclude 1 duplicate reference within set 123 24 4

mp: keyword (title, abstract, subject heading)


Table 2. EMBASE search

Search Syntax Hits Marked for

retrieval Included

1 'artificial embolism'/exp 60,172 — —

2 bronchial 93,019 — —

3 #1 AND #2 1,153 — —

4 bronchial NEAR/3 (embolization OR embolisation) 756 — —

5 #3 OR #4 1,249 — —

6 hemoptysis OR haemoptysis 18,474 — —

7 embolization OR embolization 53,920 — —

8 #6 AND #7 1,302 — —

9 #4 OR #4 OR #8 1,732 — —

10 #9 AND 'guidelines'/exp 18 — —

delete 2 references duplicating Medline results 16 6 0

11 #9 AND ([cochrane review]/lim OR [systematic review]/lim OR [meta analysis]/lim) 6 — —

delete 1 reference duplicating Medline results 5 2 0

12 #9 AND ('clinical trial'/de OR 'cohort analysis'/de OR 'comparative study'/de OR

'controlled clinical trial'/de OR 'controlled study'/de OR 'prospective study'/de) 150 — —

delete 18 references duplicating Medline results 132 33 3

Table 3. Cochrane Library search

Search Syntax Total

Hits

Cochrane

reviews DARE

Cochrane

Central

Register

HTA Marked for

retrieval Included

1 MeSH descriptor: [Embolization, Therapeutic] explode all trees 773 8 133 532 54 — —

2 bronchial 7,002 402 63 6,482 26 — —

3 #1 AND #2 1 1 0 0 0 — —

4 embolization or embolization 1,248 125 101 904 58 — —

5 #2 AND #4 7 3 0 3 0 — —

6 hemoptysis or haemoptysis 292 93 12 178 2 — —

7 #4 AND #6 5 3 0 1 0 — —

8 #3 OR #5 OR #7 8 4 0 3 0 — —

delete 2 references duplicating EMBASE results 6 4 0 1 0 0 0


Results

Guidelines

As discussed in our previous report (2), none of the published guidelines on this topic cited specific evidence as the basis for

thresholds defining massive or life-threatening hemoptysis. The searches described above found no additional relevant guidelines.

Note that many of the articles identified as guidelines in the Medline and EMBASE searches actually were narrative review articles.

Reviews

None of the reviews found in our searches provided any numeric data on the relationship between amount of hemoptysis and patient

outcomes after bronchial artery embolization. Two articles (3, 4) did provide evidence tables, which we used as a supplement to our

literature searches to locate potentially relevant primary articles. None of the articles retrieved from those references provided any

additional evidence for our tables.

Primary literature

Risk of recurrent hemoptysis

The seven studies that reported the effect of any patient characteristic on risk of rebleeding after bronchial artery embolization for

patients with massive or life-threatening hemoptysis are listed in Table 4. All of these studies were retrospective; three of the seven

were focused on tuberculosis patients while the other four studies included patients with various underlying conditions causing

hemoptysis. There was too much heterogeneity in design and execution of the studies to permit meta-analysis or other quantitative

data synthesis. None of the articles reported any criteria for selection or exclusion of this procedure other than the massive or life-

threatening hemoptysis.

None of the studies were done in the United States or other western countries. Patient characteristics and patterns of care are likely to

be different in the Asian and South African hospitals where these studies were carried out, compared to their U.S. counterparts.

Therefore the results should be used with caution when applying them to American practice.

Three studies compared hemoptysis recurrence rates between male and female patients, and between older and younger patients. No

statistically significant risk differences were found among these groups. A few studies compared patients with different underlying

diseases, including tuberculosis, lung cancer, aspergillosis, and bronchiectasis; the only significant finding was for increased

rebleeding risk for patients with aspergillosis in a single study. One other study found an increased risk of recurrence in patients with

inactive tuberculosis compared to patients with active tuberculosis.


There appears to be an increased risk of hemoptysis recurrence after BAE in patients with higher volume of hemoptysis, though one of

the three studies measuring this variable did not find a significant effect. History of previous hemoptysis was associated with

increased risk of recurrence in two studies, but two studies examining the effect of previous embolization procedures found no

significant recurrence risk increase. The number of vessels involved or embolized does not appear to have a significant effect on

recurrence risk. The presence of a bronchio-pulmonary shunt may be associated with increased recurrence, as is the presence of

pleural thickening. Comorbidity in general was not a risk factor for recurrence.

Chung et al. (5) found significant associations between several patient variables and risk of hemoptysis recurrence, but there are

apparent discrepancies in the odds ratios and confidence intervals presented in their results table. We sought clarification from the

authors but did not receive a response. Therefore we will consider the results of this study to be of high risk for bias in our GRADE

analysis of the evidence (while we do not have reason to believe that the findings are biased, this GRADE category applies to studies

where there is reason to believe there is a risk that published findings do not represent the actual results of a study).

Criteria in these studies for defining hemoptysis as massive or life-threatening varied. Mal’s definition of life-threatening hemoptysis

(6) was unusual: loss of 200 ml or more blood per hour, loss of 50 ml or more per hour in patients with chronic respiratory failure, or

more than two episodes of moderate hemoptysis (30 ml or more) within 24 hours despite the use of IV vasopressin. This paper did not

report the association of any patient variables with recurrent hemoptysis so it is not included in the main evidence table.

Garcia-Olivé et al (7, 8) reported on patient characteristics and time to recurrence in patients requiring a second embolization, but did

not report the effect of these characteristics on risk of recurrence.

Table 4. Primary studies: recurrence of hemoptysis

Author Study design Patients Hemoptsys criteria Patient variable Results (95% CI) Comment

Pei 2014 China (9)

Retrospective cohort

Tuberculosis

N = 112

400 ml/24 h Inactive tuberculosis vs. active tuberculosis

Hazard ratio 2.1 (1.01-4.4) Follow-up 2-52 months, median 20

Recurrence defined as 40 ml/24 h

Age > 50

Female

Hazard ratio 2.0 (0.81-4.9)


Govind 2013 South Africa (10)


Tuberculosis

N = 107

300-600 ml/24 h or requires transfusion

HIV infection status No effect on short-term procedure success rate

Follow-up period not reported

Recurrence requiring further treatment



Anuradha 2012 India (11)


Tuberculosis

N = 58

600 ml or more (time period not specified)

Inactive tuberculosis vs. active tuberculosis

Multiple arteries embolized

Non-bronchial systemic artery collaterals

No significant increase in rebleeding rate

Follow-up 11 days-5 years, median 14 months

No factors found to be statistically significant

Age, sex, degree of hemoptysis not tested

Trend towards increased risk in pts with systemic to pulmonary venous shunts (p = 0.054)

Recurrence defined as 30 ml

Chan 2009 Hong Kong (12)


Various

N = 167

200 ml/24 h or bleeding that was life-threatening

History of hemoptysis

Incomplete embolization in initial procedure

Bronchio-pulmonary shunt




Median follow-up 6.5 years

Regression analysis

Recurrence defined as further life-threatening bleeding

Past history of embolization

Number of abnormal vessels on arteriogram

Hazard ratio NS

Hazard ratio NS

Not significantly different from 1.0

Chung 2006 Korea (5)


Various

N = 66

100 ml or more (time period not specified)

Hemoptysis > 200 ml

Bilateral lesion

Pleural thickening

Odds ratio 10.2 (10.0-50)

Odds ratio 13.3 (1.4-138)

Odds ratio 20.8 (3.4-128)

Follow-up 1-56 months, median 20


Female

Age 50 and above

Previous embolization

Lung cancer

Multiple feeding vessels

Odds ratio 0.59 (0.53-4.4)

Odds ratio 0.42 (0.34-2.4)

Odds ratio 2.0 (0.36-3.6)

Odds ratio 1.8 (0.77-11.8)

Odds ratio 2.2 (0.36-13.4)

Unexplained discrepancies in results table: see text



Kim 2006 Korea (13)


Various

N = 118

200 ml/24 h or hematocrit < 30

Aspergillosis Significant increase in rebleeding rate

Minimum 1 year follow-up

80% of pts. had massive hemoptysis

Sex

Age

Degree of hemoptysis

Tuberculosis

Bronchiectasis

Lung cancer

APACHE II score


Recurrence requiring readmission

Kim 1997 Korea (14)


Various

N = 51

400 ml/24 h or multiple episodes > 100 ml or 100 ml or more for 5 days

Hemoptysis > 400 ml/24 hrs

History of 3 or more hemoptysis events

Multiple arteries involved

Significant increase in rebleeding rate

Follow-up 7-60 months

Insufficient patients with aspergillosis for any conclusions to be drawn


Tuberculosis

Bronchiectasis


Highlighted variables denote statistically significant association with subsequent rebleeding

Other outcomes

In most of the studies found by our searches, the only clinical outcome reported by group was recurrence of hemoptysis; studies

focused on recurrence as the main outcome of interest. Pei et al. were the only investigators to report any other outcome: they found

no significant differences in short-term control of bleeding (procedure success) as a function of age, sex, or disease status (active or

inactive) in tuberculosis patients (9). While many articles reported on the number of BAE patients with complications and/or

mortality, including some articles that did not meet the inclusion criteria for the recurrence analysis above, none analyzed these

outcomes as a function of any specific patient characteristics.

There were no studies reporting results by patient group for any of the other outcomes of interest, including (but not limited to)

procedure-related bleeding, paralysis, length of stay, or cost of care.


Conclusion

There are no guidelines or systematic reviews identifying patient groups for whom bronchial artery embolization carries an increased

risk of long-term success or failure. The only evidence on comparative rates of hemoptysis recurrence came from small to medium-

sized retrospective studies, all done in Asian countries or South Africa. We summarize the results of those studies in the GRADE

table below (Table 5). Because the only evidence is from retrospective cohort studies, the strength of evidence is low at best, and very

low for comparisons where one or more downgrades had to be applied because of inconsistent results or there being only a single

study involving a particular variable.

The only variable for which there was a significant association and no downgrade of evidence was a history of previous hemoptysis,

which is associated with an increased risk of recurrence after embolization. Patient demographic variables such as sex and age do not

appear to be associated with recurrence risk, and most specific etiologies of hemoptysis (such as tuberculosis and lung cancer) do not

appear to have an association either. A single study found that patients with aspergillosis had an increased recurrence risk, but the

strength of that evidence is very low. There is inconsistent evidence (strength: very low) suggesting that patients with inactive

tuberculosis are at greater risk of recurrent hemoptysis than patients with active tuberculosis. Two of three studies that made the

comparison found that patients with high-volume hemoptysis were at greater risk of recurrence than patients with lower-volume, but

still massive or life-threatening hemoptysis, but the inconsistency of results causes us to rate the strength of that evidence as very low.

There was only one study reporting on the relationship between patient characteristics and procedure success (short term control of

bleeding) and there was no evidence for any other outcomes, including mortality, complications of BAE, or cost.

Table 5. Evidence summary and GRADE analysis

Variable Outcome Conclusion

Quantity and type of

evidence

Starting level of evidence

strength Ris

k o

f b

ias

Inco

nsi

sten

cy

Ind

irec

tnes

s

Imp

reci

sio

n

Pu

blic

atio

n b

ias

Str

on

g o

r ve

ry

stro

ng

ass

n.

Do

se-r

esp

on

se

Co

nfo

un

der

s

con

sid

ered

Final level of evidence

strength

Sex Risk of recurrence No significant effect 3 cohort Low 0 0 0 0 0 0 0 0 Low

Age Risk of recurrence No significant effect 3 cohort Low 0 0 0 0 0 0 0 0 Low

Tuberculosis vs. other causes Risk of recurrence No significant effect 2 cohort Low 0 0 0 0 0 0 0 0 Low

Inactive vs. active TB Risk of recurrence Possible increase in risk 2 cohort Low 0 –1 0 0 0 0 0 0 Very low

Lung cancer Risk of recurrence No significant effect 2 cohort Low 0 0 0 0 0 0 0 0 Low


Variable Outcome Conclusion

Quantity and type of

evidence

Starting level of evidence

strength Ris

k o

f b

ias

Inco

nsi

sten

cy

Ind

irec

tnes

s

Imp

reci

sio

n

Pu

blic

atio

n b

ias

Str

on

g o

r ve

ry

stro

ng

ass

n.

Do

se-r

esp

on

se

Co

nfo

un

der

s

con

sid

ered

Final level of evidence

strength

Aspergillosis Risk of recurrence Increased risk 1 cohort Low 0 0 0 –1 0 0 0 0 Very low

Bronchiectasis Risk of recurrence No significant effect 2 cohort Low 0 0 0 0 0 0 0 0 Low

HIV Risk of recurrence No significant effect 1 cohort Low 0 0 0 –1 0 0 0 0 Very low

Other comorbidity (APACHE) Risk of recurrence No significant effect 1 cohort Low 0 0 0 –1 0 0 0 0 Very low

†–Very severe hemoptysis Risk of recurrence Increased risk 3 cohort Low 0 –1 0 0 0 0 0 0 Very low

Number of vessels involved Risk of recurrence No significant effect 4 cohort Low 0 –1 0 0 0 0 0 0 Very low

History of prev. hemoptysis Risk of recurrence Increased risk 2 cohort Low 0 0 0 0 0 0 0 0 Low

History of prev. embolization Risk of recurrence No significant effect 2 cohort Low 0 0 0 0 0 0 0 0 Low

Bronchio-pulmonary shunt or other collateral vessel

Risk of recurrence Possible increase in risk 2 cohort Low 0 –1 0 0 0 0 0 0 Very low

Pleural thickening Risk of recurrence Increased risk 1 cohort Low –1 0 0 –1 0 0 0 0 Very low

Patient age or sex Short term control of bleeding

No significant effect 1 cohort Low 0 0 0 –1 0 0 0 0 Very low

All variables Complications No evidence None None

All variables Mortality No evidence None None

All variables All other outcomes No evidence None None

†–At least 200 ml/event or 400 ml/24 hours.


References 1. Lavenberg JG, Umscheid CA, Mull N. Algorithms for the management of massive hemoptysis in the acute care setting. Annotated

Bibliography. Philadelphia: University of Pennsylvania Health System; 2015 January. Report No. 288.

2. Mitchell MD, Mull N, Umscheid CA, Williams K. Guidelines for management of massive hemoptysis. Evidence Advisory.

Philadelphia: University of Pennsylvania Health System; 2015 July. Report No. 299.

3. Ashleigh RJ, Webb AK. Radiological intervention for haemoptysis in cystic fibrosis. J R Soc Med. 2007;100 Suppl 47:38-45.

4. Swanson KL, Johnson CM, Prakash UBS, McKusick MA, Andrews JC, Stanson AW. Bronchial artery embolization: experience

with 54 patients. Chest. 2002;121(3):789-95.

5. Chung WY, Byun MK, Park MS, Hahn CH, Kang SM, Lee DY, et al. Risk factors of recurrent hemoptysis after bronchial artery

embolization. Tuberc Respir Dis. 2006;60(1):65-71.

6. Mal H, Rullon I, Mellot F, Brugiere O, Sleiman C, Menu Y, et al. Immediate and long-term results of bronchial artery embolization

for life-threatening hemoptysis. Chest. 1999;115(4):996-1001.

7. Garcia-Olive I, Sanz-Santos J, Centeno C, Radua J, Andreo F, Sampere J, et al. Predictors of recanalization in patients with life-

threatening hemoptysis requiring artery embolization. Arch Bronconeumol. 2014;50(2):51-6.

8. Garcia-Olive I, Sanz-Santos J, Centeno C, Andreo F, Munoz-Ferrer A, Serra P, et al. Results of bronchial artery embolization for

the treatment of hemoptysis caused by neoplasm. J Vasc Interv Radiol. 2014;25(2):221-8.

9. Pei R, Zhou Y, Wang G, Wang H, Huang X, Yan X, et al. Outcomes of bronchial artery embolization for life-threatening

hemoptysis secondary to tuberculosis. PLoS ONE. 2014;9(12).

10. Govind M, Maharajh J. The impact of coinfection with human immunodeficiency virus and pulmonary tuberculosis on the success

of bronchial artery embolisation. Br J Radiol. 2013;86(1021):20120256.

11. Anuradha C., Shyamkumar N.K., Vinu M., Surendra BN, Christopher D.J. Outcomes of bronchial artery embolization for life-

threatening hemoptysis due to tuberculosis and post-tuberculosis sequelae. Diagn Intervention Radiol. 2012;18(1):96-101.

12. Chan VL, So LKY, Lam JYM, Lau K, Chan C, Lin AWN, et al. Major haemoptysis in Hong Kong: aetiologies, angiographic

findings and outcomes of bronchial artery embolisation. Int J Tuberc Lung Dis. 2009;13(9):1167-73.

13. Kim YG, Yoon H, Ko GY, Lim C, Kim WD, Koh Y. Long-term effect of bronchial artery embolization in Korean patients with

haemoptysis. Respirology. 2006;11(6):776-81.

14. Kim KJ, Yoo JH, Sung NC, Won HS, Yoou KH, Kang HM. The factors related to recurrence after transcatheter arterial

embolization for the treatment of hemoptysis. Korean J Intern Med. 1997;12(1):45-51.


Appendix. GRADE criteria for rating a body of evidence on an intervention

Developed by the GRADE Working Group (www.gradeworkinggroup.org)

Grades and interpretations:

High: Further research is very unlikely to change our confidence in the estimate of effect.

Moderate: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.

Low: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.

Very low: Any estimate of effect is very uncertain.

Type of evidence and starting level

Randomized trial–high

Observational study–low

Any other evidence–very low

Criteria for increasing or decreasing level

Reductions

Study quality has serious (–1) or very serious (–2) problems

Important inconsistency in evidence (–1)

Directness is somewhat (–1) or seriously (–2) uncertain

Sparse or imprecise data (–1)

Reporting bias highly probable (–1)

Increases

Evidence of association† strong (+1) or very strong (+2)

Dose-response gradient evident (+1)

All plausible confounders would reduce the effect (+1)

†Strong association defined as significant relative risk (factor of 2) based on consistent evidence from two or more studies with no plausible confounders Very strong association defined as significant relative risk (factor of 5) based on direct evidence with no threats to validity.

atient characteristics associated with bronchial artery ...pennir.com/content/bronchial artery embo...

Documents