Perioperative outcomes and amputation-free
survival after lower extremity bypass surgery in
California hospitals, 1996-1999, with follow-up
Joe Feinglass, PhD,a,bMin-Woong Sohn, PhD,b,cHeron Rodriguez, MD,dGary J. Martin, MD,aand
William H. Pearce, MD,dChicago and Hines, Ill
Objective: There are very few population-based studies of long-term outcomes after lower extremity (LE) bypass graft
surgery. This study analyzes perioperative and long-term limb salvage and amputation-free survival outcomes for patients
undergoing initial aortoiliac or femoropopliteal bypass graft surgery in California hospitals from 1996 to 1999.
Methods: Administrative data with encrypted identifiers were used to identify a chronologically first, index admission of all
patients undergoing LE bypass procedures for occlusive disease from 1996 to 1999. A 1993 to 1995 look-back period
was used to exclude patients who had undergone prior bypass surgery or amputation procedures. Patients with incident
procedures were then followed forward to determine subsequent hospitalizations and vital status through 2004. The
study comprised 28,128 patients discharged from 345 California hospitals with a median 61.5-month follow-up. Risk
factors included demographic characteristics, comorbid conditions, admission type, gangrene or ulceration, operation
level, hospital LE bypass surgery volume, and year of discharge. Logistic regression was used to analyze 30-day outcomes,
and Kaplan-Meier analysis and Cox proportional hazard models were used to analyze amputation-free survival.
Results: Overall 30-day mortality was 4.3, and the 30-day major amputation rate was 2.6%. Limb salvage was 81.9% at 5
years and 76.4% at 9 years. Amputation-free survival was 51.5% at 5 years and 34.1% at 9 years. Risk factors were generally
similar for both perioperative and late outcomes. Advanced age, higher comorbidity level, gangrene, and emergency or
nursing home admission conferred significantly greater risk. Hospital volume was associated with both perioperative and
late outcomes. African American and Hispanic patients had much higher amputation rates but did not have higher
mortality risk after controlling for baseline severity of illness.
Conclusions: Long-term outcomes of LE bypass surgery were superior for high-volume hospital patients. Graft surveil-
lance and risk factor follow-up care provide a major opportunity for quality improvement efforts. The contrast between
traditional limb salvage and amputation-free survival outcomes raises questions about the value of surgical treatment,
particularly for patients with limited life expectancy and without coding of tissue loss or critical limb ischemia. (J Vasc
By the mid-1990s, meta-reviews of lower extremity
(LE) bypass surgery, based mainly on single-center studies,
after femoropopliteal vein bypass for disabling claudication
and 66% for critical limb ischemia (CLI), with approxi-
mately 15% to 20% lower rates for prosthetic grafts.1Com-
pared with primary patency, even better results have been
reported for long-term limb salvage, although often
achieved after frequent repeat revascularizations, rehospi-
talizations, and intensive local wound care.2,3Rarely, how-
ever, have large population-based datasets been used to
document rates of limb salvage over a long follow-up
period, including a full description of the burden of mor-
bidity and mortality after an initial surgery.4,5
From the patient perspective, a key long-term outcome
of LE bypass surgery is amputation-free survival, defined as
the time between a patient’s initial bypass surgery oper-
ation and either subsequent major amputation or death.
This was originally termed “cumulative palliation” and is
a benchmark for clinical decisions about which patients
will benefit most from LE bypass surgery.6Because
amputation and death are both considered failure end
points, amputation-free survival rates will be significantly
lower than crude limb salvage rates computed when
deaths are right censored.
Recent conservative benchmarks include 3-year
amputation-free survival of 57% among 195 surgically
treated patients with critical limb ischemia in the Bypass
Versus Angioplasty in Severe Ischemia of the Leg (BASIL)
trial, three-quarters of whom received vein grafts.7The
amputation-free survival rate for the 404 LE vein bypass
patients in the PREVENT III study of edofoligide for the
Division of General Internal Medicine,aInstitute for Healthcare Studies,b
and Division of Vascular Surgery,dNorthwestern University Feinberg
School of Medicine, Chicago; and Center for Management of Complex
Chronic Care, Hines VA Hospital, Hines.c
The Barney Family Foundation provided financial support for data acquisi-
tion for this project.
Competition of interest: none.
Additional material for this article may be found online at www.jvascsurg.org.
Correspondence: Joe Feinglass, PhD, General Internal Medicine, 750 N.
Lakeshore Dr, 10th Flr, Chicago, IL 60611 (e-mail: j-feinglass@
Copyright © 2009 by the Society for Vascular Surgery.
prevention of infrainguinal graft failure, which included
patients with very high rates of tissue loss and diabetes, was
74.1% at 1 year.8
This study used State of California hospital discharge
data to investigate amputation-free survival for a cohort of
?28,000 unique patients who underwent bypass proce-
dures in nonfederal hospitals in California from 1996 to
1999. Cohort inception was based on a chronologically
first index or incident LE bypass surgery determined with
the aid of a 3-year look-back period (1993-1995) to detect
Each patient in this cohort was then followed for up to 9
years to document subsequent LE bypass procedures or
amputations, as well as vital status, through 2004.
We present 30-day mortality and major complication
rates, and the frequency of subsequent rehospitalization for
bypass surgery, bypass graft revision, or major amputation
during a median 62-month follow-up period. We explore
the association of perioperative and late survival outcomes,
including limb salvage and amputation-free survival, with
patient demographic, clinical, and hospital-level risk factors
at the time of the initial surgery. Our results present a
unique picture of the epidemiology and outcomes of LE
bypass surgery in the period just before the recent rapid
growth of endovascular treatment and use of LE stents.9
Data source and selection of index admission records.
The California Office of Statewide Health Planning and
Development (OSHPD) provided patient discharge data
for the study. As part of an approved research protocol,
OSHPD provided an encrypted patient identifier to link
multiple discharges for the same patient over time. The
sample was created to select all unique patients, aged
?35 years, undergoing a first aortoiliac-femoral (AI) or
femoropopliteal-distal (FP) bypass graft surgery based on
International Classification of Disease (ICD-9) procedure
codes 39.25 or 39.29. We selected each patient’s chrono-
logically first bypass surgery admission and retained a sam-
ple of these index bypass surgery discharges of unique
We excluded patients with a principal diagnosis of aortic
anuerysm to achieve a sample of patients whose hospital
procedures were primarily related to occlusive disease.
Look-back to exclude patients with prior bypass
surgery or amputation. We then sought to exclude from
our analysis any patients with a documented history of LE
bypass surgery or amputation in California hospitals in the
3 years before cohort inception, 1993 to 1995. We identi-
fied and excluded any cohort patient with any previous
admission for through foot, below knee, or above knee
amputation, prior AI or FP procedures, or any prior bypass
graft revision procedures (ICD-9 code 39.49) in 1993 to
The look-back exlusion was designed to retain the
largest feasible number of first-time, incident LE bypass
procedures as index admissions in our analysis sample. We
did not exclude patients who only underwent prior LE
angioplasty, partly because angioplasty ICD-9 procedure
codes do not distinguish procedures anatomically. Finally,
we also excluded patients aged ?35 years.
On the basis of prior literature, we expected that ex-
cluding patients found to have previous bypass surgery or
amputation detected during a 3-year period would yield
very high rates of initial bypass procedures among the
remaining cohort.10We performed an empirical test of the
sensitivity of this 3-year look-back period by examining
through 2004. By looking at readmission rates during
3-year windows during this follow-up period, we calculated
that ?6% of our patient cohort would have had undetected
lower extremity procedures performed ?3 years after their
index operation. Because almost all study patients had an
even longer look-back than 3 years, we were reasonably
certain that our data reflected patients’ first LE bypass graft
Risk factors for perioperative and late outcome
measures. The OHSPD patient files provide information
on age and sex and whether the admission was an emer-
gency or from a nursing home. Race and ethnicity were
classified as Hispanic, African American, or white/other,
the latter category including 3.1% of all patients classified as
Asian and 2.6% as other/unknown. Chronic comorbid
the Elixhauser method,11with software distributed by the
Agency for Healthcare Research and Quality.12
The Elixhauser comorbidity codes are derived from
and resource use.13We used an equally weighted count of
18 of the most relevant codes with ?1% prevalence, includ-
ing hypertension, pulmonary, liver, renal, vascular, or cere-
brovascular disease, as well as heart failure or cancer (Ap-
severity levels. When stratified by this comorbidity index,
our study cohort had a linearly decreasing study period
survival ranging from ?61% for those with only 1 Elix-
hauser comorbidity, 52.1% for those with 2, 42.0% for
small group with ?5 conditions.
We also tested whether ICD-9 coded diabetes had an
additional incremental effect on patient outcomes. Severity
of vascular disease and extent of tissue loss are poorly
documented by ICD-9 coding. However, we followed the
literature in identifying patients with gangrene or ulcer-
ation as having the most severe presentation.
Perioperative outcomes included 30-day postproce-
dural mortality and incidence of acute myocardial infarc-
tion (MI), iatrogenic stroke, or major amputation. Mortal-
ity follow-up for all study patients through December 2004
was provided by OSHPD from a link to the state’s vital
status records. Readmissions for lower extremity vascular
surgery or amputation procedures at any nonfederal Cali-
fornia hospital were also tracked through 2004. Less than
1% of all records did not have valid patient identifiers for
linkage to vital status and readmission data and were ex-
cluded from the study.
JOURNAL OF VASCULAR SURGERY
Volume 50, Number 4
Feinglass et al 777
Amputation-free survival for each patient was deter-
mined by time from the index procedure to a failure end
point of either death or major amputation (above or below
knee) before the end of follow-up on December 31, 2004.
Limb salvage was conservatively defined as time to a subse-
quent major amputation before the end of follow-up, cen-
soring death. This is a conservative estimate, because hos-
pital administrative data do not distinguish which leg was
amputated. Because it is impossible to classify whether
subsequent amputations were for an ipsilateral or contralat-
eral limb, any major amputation was counted as evidence of
failure for the initial bypass procedure.
Hospital LE bypass surgery volume. We were able
to compute the cumulative volume of initial LE bypass
surgery at each of the 345 California hospitals in the sample
from 1996 to 1999. This volume estimate varied from a
single operation to ?600 patients at the highest-volume
institution, including bypass patients with aneurysm diag-
noses. Although these volume data were limited to the
procedures rather than total vascular surgery procedures
performed, our data nevertheless likely provided a good
measure of relative hospital vascular surgery capacity.14
We empirically categorized hospitals into four annual
LE bypass graft surgery volume strata. Low volume in-
cluded patients discharged from 244 hospitals with a 4-year
total of ?120 patients (?40 annually), medium volume
included 49 hospitals with 120 to 179 patients, high vol-
ume included 43 hospitals with 180 to 320, and very high
volume included nine hospitals with ?320 bypass graft
Statistical analysis. The significance of bivariate as-
sociations between patient demographic, clinical, and
hospital-level characteristics and 30-day outcomes were
analyzed with ?2tests. Random effects multiple logistic
regression with hospital as the class variable was used to
estimate the likelihood of 30-day death or major complica-
tions, with results adjusted simultaneously for year of op-
eration, patient demographics (age, sex, race, and ethnic-
ity), clinical characteristics (emergency admission, nursing
home admission, Elixhauser comorbidity count, diabetes,
ulceration, gangrene, AI vs FP operation), and hospital
volume level. Long-term amputation-free survival and limb
salvage outcomes were estimated with the Kaplan-Meier
procedure for bivariate comparisons and with Cox propor-
tional hazards models for all variables simultaneously. Stan-
dard errors for Cox regression models were adjusted for the
clustering of patients within hospitals using Huber-White
sandwich estimators. All statistical analyses used Stata 10
software (StataCorp, College Station, Tex).
Rates of prior LE procedures from 1993 to 1995
look-back. The 1996 to 1999 discharges with LE bypass
surgery codes initially included 36,972 unique patients.
When checked against 1993 to 1995 look-back data, 3730
(10.1%) had prior LE bypass or bypass graft revision pro-
cedures between January 1993 and the index procedure in
1996 to 1999. These patients also had 1215 (3.3%) docu-
mented prior through foot, below knee, or above knee
amputations. These exclusions, removal of patients aged
?35 years (0.9%), and ?3000 patients (8.3%) who under-
went AI procedures but had a principal diagnosis of aortic
aneurysm, resulted in a final index admission sample of
28,128 patients who underwent initial LE bypass proce-
dures at 345 California hospitals between 1996 and 1999.
Operation level, indications, and overall survival.
There were 23,014 patients (81.8%) with initial FP proce-
5114 (18.2%) with initial AI procedures. The median fol-
low-up was 61.5 months, and 47.0% of study patients had
survived to the end of follow-up in December 2004.
Kaplan-Meier survival probabilities were 80.5% at 1 year,
54.5% at 5 years, and only 38.0% at 9 years (all standard
error ?0.004). The number of new patients undergoing
initial LE bypass procedures declined slightly across the
study period, ranging from a high of 7096 in 1996 to 6997
Perioperative outcomes. Table I presents periopera-
tive mortality and major complication rates by key sample
characteristics. The overall 30-day mortality rate was 4.3%
Rates at 30-days were 2.3% overall for inpatient MI or
iatrogenic stroke, 2.6% overall for below or above knee
amputation (3.0% for FP patients; 1.0% for AI patients),
and 1.4% for FP but only 0.2% for AI patients for through
foot amputation (data not shown).
Mortality rates increased significantly by age. White/
other and Hispanic patients had higher 30-day mortality,
but African-American and Hispanic patients had more than
twice the 30-day major amputation rate. Significantly
higher 30-day mortality differences were evident by admis-
sion status (emergency or nursing home), comorbidity
counts exceeding 2, and for patients with gangrene. Mor-
tality varied inversely by hospital volume level. The nine
highest volume hospitals had a much lower 30-day mortal-
ity (3.0%) compared with the 244 low-volume institutions
(5.1%). The trend towards lower mortality by year was not
Readmission rates for subsequent bypass surgery or
major amputation. During the follow-up period, 8188
patients (29.1%) were readmitted for at least one (and up to
19) subsequent bypass graft surgery or revision procedures,
including 31.4% of FP and 18.8% of AI patients. Many
more patients were readmitted for an additional FP proce-
dure (21.9%) than for a subsequent AI procedure (2.5%),
which explains the much higher FP/AI ratio derived from
studies based on admissions rather than patients. Alto-
gether, 4796 of study patients (17.1%) underwent major
of FP and 5.3% of AI patients (P ? .0001). Major ampu-
tation rates were much higher among patients undergoing
subsequent bypass or revision procedures, comprising
29.9% of readmitted FP and 15.6% of readmitted AI pa-
tients (P ? .0001).
JOURNAL OF VASCULAR SURGERY
778 Feinglass et al
Major amputation-free survival and limb salvage.
Kaplan-Meier amputation-free survival estimates for all pa-
tients were 78.1% at 1 year, 62.9% at 3 years, 51.1% at 5
years, and 34.1% at 9 years. In contrast, limb salvage
calculated with deaths censored was 90.7% at 1 year, 85.5%
at 3 years, 81.9% at 5 years, and 76.4% at 9 years (all
standard error ?0.004). Fig 1 displays amputation-free
survival probabilities by operation level for subsequent
major amputations. There was a ?20% difference by initial
operation level by the end of the follow-up period. Fig 2
presents survival probabilities for limb salvage, indicating a
widening gap between FP and AI patient amputation rates
over time, with FP patients dipping below a 75% rate at
about 8 years. Fig 3 presents limb salvage rates by race and
ethnicity. Limb salvage probabilities for African American
and Hispanic patients dipped below 75% after about 3
years, whereas white/other patients remained well above
Multivariate risk factors for 30-day mortality and
amputation-free survival. Table II presents odds ratios
derived from the logistic regression model and hazard
ratios derived from the Cox model for patient risk factors.
Confidence intervals that exclude 1.0 are statistically signif-
icant. The logistic model results for 30-day mortality gen-
erally mirror Table I but provide some additional insights,
especially when compared with Cox model results the next
column. As expected, older patients were at higher risk.
Men did not differ significantly at 30 days but had worse
Table I. Perioperative mortality and complications after lower extremity bypass surgery from 1996 to 1999 in 28,128
index admissionsato 345 California hospitals
VariablesPatients (% of sample) Inpatient MI or stroke, %
30-day postprocedure, %
Mortality Major amputation
Race and ethnicityb
Nursing home admit
Elixhauser comorbidities, No.b
Hospital LE bypass surgery volumeb
Very high, ?80/y
Year of discharge
LE, Lower extremity; MI, myocardial infarction.
aThe index admission sample was all unique patients undergoing lower extremity bypass in 1996 to 1999 without having any previous admissions for bypass
surgery, bypass graft revision, or amputation between 1993 and 1995.
bThe significance of 30-day mortality differences was P ? .001.
JOURNAL OF VASCULAR SURGERY
Volume 50, Number 4
Feinglass et al 779
amputation-free survival. African American race was actu-
ally protective compared with white/other at 30 days, and
race and ethnicity differences were not significant for
amputation-free survival. FP surgery was 47% less risky at
30 days but conferred a 27% greater risk for amputation-
Increasing comorbidity levels were associated with in-
ferior perioperative and late outcomes. Diabetes had no
for late survival. Emergency and nursing home admissions
and gangrene were highly significant risk factors, but cod-
ing of LE ulcer was only significant for late outcomes.
Hospital volume was significant for both early and late
outcomes. The 1999 to 1996 comparison was significant
for declining 30-day mortality only. Year of surgery was not
a significant predictor of amputation-free survival.
To our knowledge, this study provides the longest
follow-up period of any population-based study of LE
bypass surgery to date. The only other population-based
we are aware of, was the report by Al-Omran et al4on 1991
to 1998 data from Ontario, Canada. This study used some-
what different terminology and reported higher 5-year
overall patient survival (61.5%), but a similar 5-year limb
salvage rate (83.4%). However, the Ontario study was
primarily concerned with comparisons with endovascular
results. The authors did not report perioperative outcomes
or a combined amputation or death end point, did not use
a look-back procedure to identify index initial operations,
and did not examine the same risk factors.
Outcome findings by hospital volume. In particular,
our study also found that the strong association between
hospital volume and perioperative death15extends to late
amputation-free survival. This implies that hospital quality
differences may extend to follow-up care and graft surveil-
lance. Indeed, the effect of hospital system quality of care
on late survival outcomes may affect many more lives than
early postoperative quality differences. Follow-up care is
thus a potentially more rewarding target for quality im-
provement than operating room care.16
Race and ethnicity findings. Also of interest was that
after controlling for age and baseline health factors, such as
comorbidity and tissue loss coding, and despite signifi-
cantly higher rates of major amputation, minority patients
actually had better perioperative and essentially identical
amputation-free survival outcomes as white/other race pa-
tients. This discrepancy between amputation and death has
Fig 1. Kaplan-Meieramputation-freesurvivalprobabilitiesbyby-
pass surgery level in 5114 patients undergoing aortoiliac (dashed
line) and 23,014 undergoing femoropopliteal bypass (solid line) at
345 California hospitals from 1996 to 1999, with follow-up
through 2004 (P ? .0001).
Fig 2. Kaplan-Meier limb salvage probabilities by bypass surgery
level in 5114 patients undergoing aortoiliac (dashed line) and
23,014 undergoing femoropopliteal bypass (solid line) at 345
California hospitals from 1996 to 1999, with follow-up through
2004 (P ? .0001).
Fig 3. Kaplan-Meier limb salvage probabilities by race and eth-
nicity for 21,934 white/other race patients (solid line) 2569 Afri-
can American patients (large dashes), and 3625 Hispanic patients
(small dashes) undergoing aortoiliac or femoropopliteal bypass
(N ? 28,128) at 345 California hospitals from 1996 to 1999, with
follow-up through 2004 (P ? .0001).
JOURNAL OF VASCULAR SURGERY
780 Feinglass et al
patients had more than twice the prevalence of gangrene
codes at the initial surgery (34.1% and 36.7% to 19.3%
respectively) and were over-represented among those with
the highest Elixhauser comorbidity counts (23.8% of Afri-
can Americans and 22.1% of Hispanics had ?4 comorbid
conditions vs 14.8% of whites). Thus, although minority
patients are known to have higher per capita rates of pe-
ripheral vascular disease and amputation,19-21there was no
disparity in survival outcomes among bypass surgery pa-
tients in California after controlling for baseline risk.
When is LE bypass surgery merited? The California
and Ontario studies raise important questions about the
well-known divergence between more traditional limb sal-
vage and amputation-free survival results. This discrepancy
has led some to question the benefit of aggressive surgical
treatment of late-stage peripheral vascular occlusive disease
in the context of the limited life expectancy of many pa-
tients. In a critical review, Nehler et al22describe overly
aggressive limb salvage as “lesion-focused” rather than
“patient-focused” care. They question whether surgical
revascularization procedures are better than primary ampu-
tation for many patients given limited functional benefits
and the potentially iatrogenic effect of interventional man-
agement on the progression of systemic atherosclerosis.
Goshima et al2also question traditional limb salvage out-
comes for critical limb ischemia because they may conceal
the major burden of subsequent morbidity in even “suc-
cessful” limb salvage.
The data presented here also provide evidence of high
30-day mortality rates that exceed estimates based solely on
inpatient deaths.23,24This raises further concerns when
Table II. Logistic regression results for 30-day postprocedural mortality and Cox proportional hazards results for
amputation-free survival in 28,128 index admissionsato 345 California hospitals
30-day postprocedural mortality
OR (95% CI)
Amputation-free survival over median
HR (95% CI)
Race and ethnicity
Nursing home admit
Elixhauser comorbidities, No.
Hospital LE bypass surgery volume
Very high, ?80/y
Year of discharge
CI, Confidence interval; HR, hazard ratio; LE, lower extremity; OR, odds ratio.
bypass graft revision, or amputation between 1993 and 1995.
JOURNAL OF VASCULAR SURGERY
Volume 50, Number 4
Feinglass et al 781
tissue loss or rest pain diagnoses are poorly documented or
15.6% had ulceration codes, only 10.5% had rest pain and
4.8% had claudication codes, including 11.0% of AI pa-
tients. A large proportion of study patients had generic
principal diagnoses of complications of diabetes, other pe-
ripheral vascular disease, LE thrombosis, or arterial stric-
about whether high perioperative death and an increased
risk for subsequent limb-threatening graft failure were ap-
propriate risks for many patients in the sample. The recent
shift to endoluminal interventions, which now use im-
proved guidewire and balloon and stent technologies, no
doubt reflects improved risk stratification, especially for
On the other hand, the high initial mortality and rehos-
pitalization rates described here should be placed in the
context of equally high rates of mortality and subsequent
morbidity after above or below knee amputation.26In this
study, 9974 of all patients (31.4%) were alive 5 years after
their initial operation without major amputation or subse-
quent bypass or revision procedures. Furthermore, limited
data evaluating functional status and quality of life after LE
bypass indicate that many survivors do maintain indepen-
dent living and walking ability and better quality of
life.27-30Thus medical decision making for many patients,
especially those with extensive gangrene and advanced re-
nal or heart failure, remains daunting.
Increase in endoluminal therapies and decline of
bypass surgery. In this context, recent reports on endolu-
minal interventions have argued that newer LE angioplasty
and stent technologies achieve equivalent or better second-
ary patency and limb salvage results than bypass graft
surgery.31-33Detailed recommendations for best practices
have recently been published by Beard,34with the caveat
that “the lack of published or ongoing randomized con-
trolled trials [is] both surprising and somewhat embarrass-
ing.” Surgical LE bypass procedures have substantially de-
clined nationally, and a large corresponding increase has
occurred in endoluminal procedures.24National estimates
indicate that ?75,000 FP and ?16,000 AI procedures
were performed in 2006; down 25% and 35%, respectively,
from 2000.35However, it remains unclear the extent to
which this decline is due to substitution of endoluminal
procedures or to a decline in critical limb ischemia preva-
lence related to improved care for diabetes and more wide-
spread use of statins.19,36The major amputation rate in the
United States declined an estimated 14% between 2000
Limitations. This study is limited by use of adminis-
trative data which do not enable us to assess more detailed
clinical characteristics, including specific bypass procedures
grouped by the two generic ICD-9 codes, or the common
subsequent treatment of contralateral limbs.37We do not
hospitals, which may perform 5% to 10% of all LE bypass
operations for men nationally.38We do not assess the
contribution of endoluminal interventions to limb salvage
outcomes or their concurrent use in our sample. Perhaps
most importantly, we cannot assess the contribution of LE
bypass surgery to reducing the overall amputation rate in
California. Because the rate of primary amputation (with-
out attempted revascularization) is unknown, the effect of
LE bypass surgery or endoluminal therapies on reducing
the amputation rate will remain obscure and controversial.
Northwestern medical student Sean Wang contributed
to this project as part of a Lifeline Foundation summer
research internship, and Elissa Oh of the Institute for
Healthcare Studies provided important research assistance.
Conception and design: JF, WP, MS, HR, GM
Analysis and interpretation: JF, MS, WP, HR, GM
Data collection: MS
Writing the article: JF
Critical revision of the article: JF, WP
Final approval of the article: JF, WP, GM, HR, MS
Statistical analysis: JF, MS
Obtained funding: WP
Overall responsibility: JF
1. Hunink MG, Wong JB, Donaldson MC, Meyerovitz MF, Harrington
DP. Patency results of percutaneous and surgical revascularization for
femoropopliteal arterial disease. Med Decis Making 1994;14:71-81.
2. Goshima KR, Mills JL Sr, Hughes JD. A new look at outcomes after
infrainguinal bypass surgery: traditional reporting standards systemati-
cally underestimate the expenditure of effort required to attain limb
salvage. J Vasc Surg 2004;39:330-5.
3. Nicoloff AD, Taylor LM Jr, McLafferty RB, Moneta GL, Porter JM.
Patient recovery after infrainguinal bypass grafting for limb salvage. J
Vasc Surg 1998;27:256-63; discussion 264-6.
4. Al-Omran M, Tu JV, Johnston KW, Mamdani MM, Kucey DS. Use of
interventional procedures for peripheral arterial occlusive disease in
5. Feinglass J, Pearce WH, Martin GJ, Gibbs J, Cowper D, Sorensen M, et
al. Postoperative and amputation-free survival outcomes after femoro-
Affairs National Surgical Quality Improvement Program. J Vasc Surg
6. Sladen JG, Gilmour JL, Wong RW. Cumulative patency and actual
palliation in patients with claudication after aortofemoral bypass. Pro-
spective long-term follow-up of 100 patients. Am J Surg 1986;152:
7. Adam DJ, Beard JD, Cleveland T, et al. Bypass versus angioplasty in
severe ischaemia of the leg (BASIL): multicentre, randomised con-
trolled trial. Lancet 2005;366:1925-34.
al. Results of PREVENT III: a multicenter, randomized trial of edifo-
ligide for the prevention of vein graft failure in lower extremity bypass
surgery. J Vasc Surg 2006;43:742-51; discussion 751.
9. Anderson PL, Gelijns A, Moskowitz A, Arons R, Gupta L, Weinberg A,
et al. Understanding trends in inpatient surgical volume: vascular inter-
ventions, 1980-2000. J Vasc Surg 2004;39:1200-8.
10. Sambamoorthi U, Tseng CL, Rajan M, Anjali T, Findley PA, Pogach L.
Initial nontraumatic lower-extremity amputations among veterans with
diabetes. Med Care 2006;44:779-87.
11. Elixhauser A, Steiner C, Harris DR, Coffey RM. Comorbidity measures
for use with administrative data. Med Care 1998;36:8-27.
JOURNAL OF VASCULAR SURGERY
782 Feinglass et al
12. Agency for Healthcare Research and Quality. Hospital Cost and Utili-
zation Project (HCUP). Tools and software. http://www.ahrq.gov/
data/hcup/#tools. Accessed: Mar 21, 2009.
13. Southern DA, Quan H, Ghali WA. Comparison of the Elixhauser and
Charlson/Deyo methods of comorbidity measurement in administra-
tive data. Med Care 2004;42:355-60.
14. Ebaugh JL, Feinglass J, Pearce WH. The effect of hospital vascular
procedures. Surgery 2001;130:561-7; discussion 567-9.
15. Killeen SD, Andrews EJ, Redmond HP, Fulton GJ. Provider volume
and outcomes for abdominal aortic aneurysm repair, carotid endarter-
ectomy, and lower extremity revascularization procedures. J Vasc Surg
16. Bilimoria KY, Bentrem DJ, Feinglass JM, Stewart AK, Winchester DP,
Talamonti MS, et al. Directing surgical quality improvement initiatives:
comparison of perioperative mortality and long-term survival for cancer
surgery. J Clin Oncol 2008;26:4626-33.
17. Chew DK, Nguyen LL, Owens CD, Conte MS, Whittemore AD,
Gravereaux EC, et al. Comparative analysis of autogenous infraingui-
nal bypass grafts in African Americans and Caucasians: the association
of race with graft function and limb salvage. J Vasc Surg 2005;42:695-
GL, et al. Disparity in outcomes of surgical revascularization for limb
and limb loss. Circulation 2009;119:123-30.
of racial disparities in lower extremity amputation rates in Northern
Illinois, 1987-2004. J Vasc Surg 2008;47:1001-7; discussion 1007.
20. Feinglass J, Rucker-Whitaker C, Lindquist L, McCarthy WJ, Pearce
WH. Racial differences in primary and repeat lower extremity amputa-
tion: results from a multihospital study. J Vasc Surg 2005;41:823-9.
21. Morrissey NJ, Giacovelli J, Egorova N, Gelijns A, Moskowitz A,
McKinsey J, et al. Disparities in the treatment and outcomes of
vascular disease in Hispanic patients. J Vasc Surg 2007;46:971-8.
treatment for critical limb ischemia? J Cardiovasc Surg (Torino) 2004;
Hospital vascular surgery volume and procedure mortality rates in
California, 1982-1994. J Vasc Surg 1998;28:45-56; discussion 56-8.
24. Nowygrod R, Egorova N, Greco G, Anderson P, Gelijns A, Moskowitz
A, et al. Trends, complications, and mortality in peripheral vascular
surgery. J Vasc Surg 2006;43:205-16.
25. Rowe VL, Lee W, Weaver FA, Etzioni D. Patterns of treatment for
peripheral arterial disease in the United States: 1996-2005. J Vasc Surg
26. FeinglassJ,PearceWH,MartinGJ,GibbsJ,CowperD,Sorensen M, et
al. Postoperative and late survival outcomes after major amputation:
findings from the Department of Veterans Affairs National Surgical
Quality Improvement Program. Surgery 2001;130:21-9.
27. Feinglass J, McCarthy WJ, Slavensky R, Manheim LM, Martin GJ.
Functional status and walking ability after lower extremity bypass graft-
ing or angioplasty for intermittent claudication: results from a prospec-
tive outcomes study. J Vasc Surg 2000;31:93-103.
28. Holtzman J, Caldwell M, Walvatne C, Kane R. Long-term functional
status and quality of life after lower extremity revascularization. J Vasc
29. Nguyen LL, Moneta GL, Conte MS, Bandyk DF, Clowes AW, Seely
BL. Prospective multicenter study of quality of life before and after
lower extremity vein bypass in 1404 patients with critical limb ischemia.
J Vasc Surg 2006;44:977-83; discussion 983-4.
30. Pomposelli FB Jr, Arora S, Gibbons GW, Frykberg R, Smakowski P,
Campbell DR, et al. Lower extremity arterial reconstruction in the very
elderly: successful outcome preserves not only the limb but also resi-
dential status and ambulatory function. J Vasc Surg 1998;28:215-25.
31. Baril DT, Marone LK, Kim J, Go MR, Chaer RA, Rhee RY. Outcomes
of endovascular interventions for TASC II B and C femoropopliteal
lesions. J Vasc Surg 2008;48:627-33.
32. Lee LK, Kent KC. Infrainguinal occlusive disease: endovascular inter-
vention is the first line therapy. Adv Surg 2008;42:193-204.
33. Romiti M, Albers M, Brochado-Neto FC, Durazzo AE, Pereira CA, De
Luccia N. Meta-analysis of infrapopliteal angioplasty for chronic critical
limb ischemia. J Vasc Surg 2008;47:975-81.
34. Beard JD. Which is the best revascularization for critical limb ischemia:
Endovascular or open surgery? J Vasc Surg 2008;48(6 suppl):11S-16S.
35. Agency for Healthcare Research and Quality. HCUPnet Nationwide
Inpatient Sample. http://hcupnet.ahrq.gov/HCUPnet.jsp. Accessed:
Mar 21, 2009.
of cholesterol-lowering with simvastatin on peripheral vascular and
other major vascular outcomes in 20,536 people with peripheral arterial
disease and other high-risk conditions. J Vasc Surg 2007;45:645-54;
37. de Vries SO, Donaldson MC, Hunink MG. Contralateral symptoms
after unilateral intervention for peripheral occlusive disease. J Vasc Surg
38. Mayfield JA, Reiber GE, Maynard C, Czerniecki JM, Caps MT, San-
georzan BJ. Trends in lower limb amputation in the Veterans Health
Administration, 1989-1998. J Rehabil Res Devel 2000;37:23-30.
Submitted Mar 26, 2009; accepted May 21, 2009.
Additional material for this article may be found online
JOURNAL OF VASCULAR SURGERY
Volume 50, Number 4
Feinglass et al 783
Appendix I (online only). Sample prevalence of Download full-text
Elixhauser chronic disease comorbidities used to create a
counted five-level comorbidity severity index for 31,220
patients discharged from 345 California hospitals after
lower extremity bypass surgery, 1996 to 1999
Peripheral vascular disease
Without chronic complications
With chronic complications
Congestive heart failure
Chronic pulmonary disease
Other neurologic disorders
Solid tumor without metastasis
Fluid and electrolyte disorders
JOURNAL OF VASCULAR SURGERY
783.e1 Feinglass et al