Analysis of wound healing time and wound-free period as

outcomes after surgical and endovascular revascularization

for critical lower limb ischemia
Jin Okazaki, MD, PhD,a Daisuke Matsuda, MD,a Kiyoshi Tanaka, MD, PhD,a Masaru Ishida, MD, PhD,a
Sosei Kuma, MD, PhD,b Koichi Morisaki, MD, PhD,c Tadashi Furuyama, MD, PhD,c and
Yoshihiko Maehara, MD, PhD,c Kitakyushu, Koga, and Fukuoka, Japan

ABSTRACT
Background: Traditional end points, such as amputation-free survival, used to assess the clinical effectiveness of lower
limb revascularization have shortcomings because they do not account independently for wound nonhealing and
recurrence or patient survival. Wound healing process and maintenance of a wound-free state after revascularization
were not well-studied. The aim of this study was to elucidate the long-term clinical course of ischemic wounds after
revascularization. We focused on initial wound healing process as well as the maintenance of a wound-free state after
achievement of wound healing. We introduced a wound-free period (WFP; the period during which limbs maintained an
ulcer-free state) and Wound Recurrence and Amputation-free Survival (WRAFS) as parameters and tested their effec-
tiveness in evaluating clinical outcomes of limbs treated using endovascular therapy (EVT) and surgical revascularization.
Methods: The medical records of patients developing lower critical limb ischemia with tissue loss who underwent surgical
or endovascular revascularization of the infrainguinal vessels between 2009 and 2013 were reviewed retrospectively. The
risk factors for achieving wound healing and WRAFS were analyzed using Kaplan-Meier survival curves and Cox regression
model. Risk factors to prolong wound healing time (WHT) and reduce WFP were determined by the least squares method.
Results: In total, 233 patients underwent 278 limb revascularizations; 138 endovascular and 140 surgical procedures were
performed as first treatments. The proportion of healed wounds 1, 2, and 3 years after primary revascularization was
64.0%, 69.7%, and 70.5%, respectively. Significant risk factors for wound healing were an EVT-first strategy (risk ratio [RR],
2.47), congestive heart failure (RR, 2.05), and wound, ischemia, and foot infection wound grade (RR, 1.59). The mean WHT
was 143.7 days. An EVT-first strategy and wound infection contributed to significantly longer WHT. The mean WFP was
711.0 days. An EVT-first strategy, history of coronary artery disease, and dialysis dependence were associated with
significantly shorter WFPs. WRAFS at 1 and 2 years after achievement of wound healing were 76.9% and 64.2%,
respectively. Significant risk factors against WRAFS were a history of coronary artery disease (RR, 1.68), dialysis depen-
dence (RR, 2.03), and being wheel chair bound (RR, 1.64).
Conclusions: EVT revascularization was associated with longer WHT, reduced wound healing rate, and a shorter WFP
compared with surgical revascularization. wound, ischemia, and foot infection grade was associated with longer WHT
and reduced wound healing rate, but not associated with a shorter WFP. Systemic conditions such as dialysis depen-
dence, congestive heart failure, and being wheel chair bound were associated with reduced wound healing rate and
shorter WFP, presumably because they limited life expectancy. WHT and WFP are useful criteria for evaluating limb
outcomes in patients with critical limb ischemia. (J Vasc Surg 2017;-:1-9.)

Ischemic lower limb revascularization in peripheral
arterial occlusive disease has been evaluated with
various end points. Technical end points include graft
patency, restenosis, or target limb revascularization
From the Department of Vascular Surgery, Kokura Memorial Hospital,
Kitakyushua; the Department of Vascular Surgery, Fukuoka Higashi Medical
Center, Kogab; and the Department of Surgery and Science, Graduate School
of Medical Sciences, Kyushu University, Fukuoka.c
Author conflict of interest: none.
Correspondence: Jin Okazaki, MD, PhD, Department of Vascular Surgery,
Kokura Memorial Hospital, 3-2-1, Asano, Kokurakita-ku, Kitakyushu-shi,
Fukuoka 8020001, Japan (e-mail: okadoc2001@yahoo.co.jp).
The editors and reviewers of this article have no relevant financial relationships to
disclose per the JVS policy that requires reviewers to decline review of any
manuscript for which they may have a conflict of interest.
0741-5214
Copyright Ó 2017 by the Society for Vascular Surgery. Published by Elsevier Inc.
http://dx.doi.org/10.1016/j.jvs.2017.07.122
(TLR). Clinical end points include survival, limb salvage
rate, amputation-free survival (AFS), and wound healing.
Most large-scale studies of critical limb ischemia (CLI) use
AFS as a clinical end point.1-3 However, AFS is a minimum
goal and is not always the same as a positive limb
outcome. The ideal outcome of revascularization in
patients with peripheral arterial disease is complete re-
lief from ischemic symptoms, and in patients with
ischemic wounds, the primary goal is wound healing.
Therefore, clinical evaluations of these patients should
assess wound healing and the maintenance of a
wound-free state. However, only a few studies have
focused on wound healing,4 and the question as to
what kind of patients, limbs, and wounds have a risk of
delayed wound healing, and whether endovascular or
surgical revascularization (SUR) provides better limb out-
comes still needs to be answered.5-8 Even when limbs
achieve wound healing, there is another question about

1
2 Okazaki et al
how long they can maintain a wound-free state. The
long-term limb outcome after the achievement of
wound healing, that is, recurrence of ischemic wound
or maintenance of a wound-free state, has not been
well-studied. The aim of this study was to elucidate the
long-term clinical course of ischemic wounds after revas-
cularization. We focused on initial wound healing pro-
cesses as well as the maintenance of a wound-free
state after achievement of wound healing. End points
included wound healing rate, wound healing time
(WHT), and we introduced the parameter wound-free
period (WFP; defined as the period during which pa-
tients maintained a wound-free state), and examined
its usefulness as a criterion for evaluating the clinical re-
sults of revascularization.
METHODS
The study protocol was developed in accordance with
the Declaration of Helsinki, and was approved by the
Institutional Review Board of Kokura Memorial Hospital.
Informed consent for this study was obtained from
each patient.
This retrospective, single-center study was conducted
between 2009 and 2013. We included patients who
developed lower limb CLI with tissue loss, underwent
surgical or endovascular revascularization of the infrain-
guinal vessels, and could be observed for 3 years after
revascularization.
We reviewed demographic data, comorbidities,
ischemic wound conditions, the details of SUR or endo-
vascular therapy (EVT) procedures, target vessels, compli-
cations, and clinical outcomes. Examined comorbidities
included hypertension, hyperlipidemia, diabetes mellitus
(insulin-dependent or controlled by oral medication),
ischemic coronary disease (history of myocardial infarc-
tion, angina, coronary bypass, or percutaneous coronary
interventions), end-stage renal disease (ESRD), atrial
fibrillation, congestive heart failure (CHF; defined as left
ventricular ejection fraction of <40% on transthoracic
echocardiogram), and cerebrovascular disease. Ischemic
wounds were graded using the Society for Vascular
Surgery wound, ischemia, and foot infection (WIfI)
classification.
For EVT, target vessels for revascularization were
defined as femoropopliteal (FP) lesion, infrapopliteal
(below the knee [BK]), or both. For SUR, target vessels
were defined as distal anastomosis at the FP artery and
infrapopliteal arteries including paramalleolar arteries.
All the limbs and patients were observed for 3 years after
primary revascularization.
Analysis of traditional end points. Traditional technical
end points included primary and secondary patency and
TLR. Traditional clinical end points included absence of
major limb amputation (amputation of the leg above
the ankle), overall survival, and AFS.
Journal of Vascular Surgery
--- 2017
ARTICLE HIGHLIGHTS
d
Type of Research: Retrospective, nonrandomized
cohort study
d
Take Home Message: After 278 limb revasculariza-
tions for critical limb ischemia with limb wounds,
the mean wound healing time was 144 days, signifi-
cantly longer with wound infection and endovascu-
lar therapy-first strategy. The mean wound-free
period was 711 days, significantly shorter with endo-
vascular therapy-first strategy, coronary artery dis-
ease, and dialysis dependence.
d
Recommendation: Initial open surgical revasculari-
zation is better than an endovascular therapy-first
approach to heal wounds fast and keep them
healed in patients with critical limb ischemia.
Analysis of wound healing. Wound healing was
defined as the complete epithelialization of an original
ulcer or stump of a minor amputation. The percentage
of limbs that achieved wound healing was calculated
and compared between the groups. WHT was defined
as the number of days required to achieve complete
wound healing after revascularization. The relationships
between WHT and the categorical risk factor variables
were analyzed using Kaplan-Meier curves. Beyond the
Kaplan-Meier analysis, limbs that did not achieve wound
healing were excluded from the analysis of mean WHT.
Maintenance of a wound-free state. For evaluating the
ability to not only achieve but also maintain wound heal-
ing, the parameters CLI recurrence, WFP, and wound
recurrence and AFS (WRAFS) were used. The observation
period was fixed at 3 years (1095 days) after primary
revascularization. CLI recurrence was defined as the
recurrence of an ischemic wound at its original position
or the occurrence of a new wound in the ipsilateral
limb. Cumulative analysis of CLI recurrence was assessed
using Kaplan-Meier curves. WFP was defined as the
period during which limbs maintained an ulcer-free
state. WFP was calculated as the period between the
date of wound healing and the date of CLI recurrence or
the end of observation owing to death, amputation, or
confirmation of a healed wound 1095 days after revas-
cularization. Mean WFP represented the capacity of
revascularization to achieve wound healing quickly and
durably maintain wound-free status. WRAFS was the
cumulative incidence of CLI recurrence, amputation, or
death after achievement of wound healing, assessed
using Kaplan-Meier curves.
Statistical analysis focused on defining the indepen-
dent risk factors against positive clinical outcomes.
Each end point was assessed using Kaplan-Meier survival
curves. The log-rank test was used to determine whether
the curves differed across the levels of categorical risk
Journal of Vascular Surgery Okazaki et al 3
Volume -, Number -

Table I. Limb demographics

Factor All (N ¼ 278) EVT first (n ¼ 138) SUR first (n ¼ 140) P
Age, years 72.5 73.2 71.8 .2274
Male/female 178/100 83/55 95/45 .1801
Diabetes mellitus 198 (71.2) 104 (75.3) 94 (67.1) .1296
History of CVD 78 (28.6) 38 (27.5) 40 (28.5) .8477
History of CAD 160 (57.5) 81 (58.7) 79 (56.4) .7022
CHF (LVEF < 40%) 38 (14.0) 20 (15.0) 18 (13.1) .6537
Dialysis-dependent ESRD 161 (57.9) 84 (60.8) 77 (55.0) .3214
Wheel chair bound 85 (30.5) 60 (43.4) 25 (17.8) <.0001a
WIfI wound grade $ 2 191 (68.7) 90 (65.2) 101 (72.1) .2128
Wound infection 117 (42.0) 57 (41.3) 60 (42.8) .7932
Heel wound 29 (10.4) 16 (11.5) 13 (9.2) .5287
CAD, Coronary artery disease; CVD, cerebrovascular disease; ESRD, end-stage renal disease; EVT, endovascular therapy; LVEF, left ventricular ejection
fraction; SUR, surgical revascularization; WIfI, wound, ischemia, and foot infection.
Values are number (%) unless otherwise indicated.
a
Indicates statistical significance (P < .05).

Table II. Primary revascularization and repeated interventions

Primary intervention
278 Limbs 137 Limbs that required repeated or
EVT/SUR first Target artery (n ¼ 278), No. additional revascularizations (n ¼ 137)
EVT first (138 limbs) FP 55 EVT first, 74 (53.6%) EVT 39
BK 41 Surgical 35
FPþBK 42
SUR first (140 limbs) FP 35 SUR first, 63 (45%) EVT 35
Infrapopliteal 105 Surgical 28
BK, Below the knee; EVT, endovascular therapy; FP, femoropopliteal; SUR, surgical revascularization.

factors. Variables with a P value of < .1 in the univariate Table III. Patient, limb, and wound outcome at 3 years
analysis were incorporated into a multivariate Cox after revascularization
regression model to obtain risk ratios (RRs) for the factors Outcome at 3 years All, EVT, SUR,
against each end point. after revascularization No. (%) No. (%) No. (%)
Continuous data were expressed as mean 6 standard Survival
deviation. The statistical difference with or without cate- Patients 118 52 66
gorical variables was analyzed using the Student test for Limbs 140 60 80
continuous data and the c2 test for categorical data. Lost limbs 11 (7.8) 7 (11.6) 4 (5.0)
Variables with a P value of < .05 were considered signif-
Wound unhealed 9 (6.4) 7 (11.6) 2 (2.5)
icant. All analysis were conducted using JMP 10.0
Wound healed 131 (93.5) 53 (88.3) 78 (97.5)
software (SAS Institute, Cary, NC).
Death
Patients 115 63 52
Limbs 138 78 60
RESULTS
Lost limbs 28 (20.2) 19 (24.3) 9 (15.0)
A total of 233 patients underwent 278 limb revasculari-
Wound unhealed 73 (52.8) 48 (61.5) 25 (41.6)
zations. Sixty patients underwent revascularizations in
both legs. The patient and limb characteristics are shown Wound healed 65 (47.1) 30 (38.4) 35 (58.3)
in Table I. The mean patient age was 72.5 years, and 64% EVT, Endovascular therapy; SUR, surgical revascularization.
of the patients were male. A total of 58% of the patients
were dialysis dependent, 71% had diabetes mellitus, 57%
had a history of ischemic heart disease, and 30% of Among the 278 limb revascularizations, 138 endovascu-
patients were wheel chair bound. All 278 limbs had lar and 140 surgical procedures were performed as first
ischemic ulcers. Wound grades according to the WIfI treatments. One hundred thirty-seven additional or
classification were 1 (31.6%), 2 (65.1%), and 3 (3.2%). repeated revascularizations in the ipsilateral limbs (TLR)
4 Okazaki et al Journal of Vascular Surgery
--- 2017

Table IV. Cumulative achievement of wound healing at

3 years after primary revascularizationa
Limbs achieving
Factor No. wound healing, % P
All limbs 278 70.5 d
EVT-/SUR-first strategy
EVT 138 60.1 .0002b
SUR 140 80.7
Diabetes mellitus
Yes 198 68.1 .1760
No 80 76.2
History of CAD
Yes 160 67.5 .1990 1 years 2 years 3 years
No 118 74.5
Healed wound (%) 64.0 69.7 70.5
Dialysis-dependent ESRD
Limbs at risk 101 85 83
Yes 161 60.2 <.0001b
No 117 84.6
Fig 1. Kaplan-Meier analysis for wound healing. For the
CHF (LVEF < 40%)
limbs eliminated from observation by major amputation
Yes 38 50.0 .0002b or death, wound healing time (WHT) was considered to be
No 232 73.2 infinite (treated as 2000 days in the statistical analysis).
Wheel chair bound
Yes 85 52.9 <.0001b
No 193 78.2
Of the 278 limbs, 196 achieved wound healing. Overall
survival, freedom from limb loss, and wound healing
WIfI wound grade
rate at 3 years were 51.4%, 83.9%, and 70.5%, respectively
1 87 83.9 .0006b
(Table III).
$2 191 64.4
Wound infection SUR-first strategy. Of the 140 limbs in the SUR-first
Yes 117 65.8 .1449 group, 7 underwent thromboendarterectomy, 24 above-
No 161 73.9 the-knee FP bypasses, 4 BK bypasses, and 105 infrapo-
Heel wound
pliteal (perimalleolar) arterial bypasses. Target vessel
locations included 35 FP and 105 infrapopliteal vessels.
Yes 29 58.6 .1092
Four of the 140 SURs (4.6%) failed within 30 days and two
No 249 71.8
of the limbs underwent early reoperation. Two reopera-
CAD, Coronary artery disease; CHF, congestive heart failure; ESRD, end-
stage renal disease; EVT, endovascular therapy; LVEF, left ventricular
tion contained one thrombectomy and one patch an-
ejection fraction; SUR, surgical revascularization; WIfI, wound, ischemia, gioplasty on distal anastomosis. Of the 140 limbs, 63 (45%)
and foot infection.
a
An EVT-first strategy, dialysis dependence, chronic heart failure, being
required TLRs because of graft failure or because primary
wheel chair bound, and a WIfI wound grade $ 2 had significantly revascularization was considered hemodynamically un-
reduced wound healing rate.
b
Indicates statistical significance (P < .05).
successful. These TLRs included 28 surgical revisions and
35 EVT procedures. Thirteen major amputations were
performed within the observation period, and 113 of 140
the limbs (80.7%) achieved wound healing. Of 113 limbs
were performed owing to failed or ineffective primary
that achieved wound healing, 14 had recurrent ischemic
revascularization. TLRs included repeated revasculariza-
ulcer and three were amputated owing to recurrent CLI.
tion for restenosis of primary treatment, and additional
Of the 27 limbs that did not achieve wound healing,
revascularization for diseases at upstream or down-
wound observation was terminated by amputation in 10
stream from primary treatment site (Table II).
cases and by death in 17 cases (Table II).
At 3 years after primary revascularization, 115 of the 233
patients had died, and 138 of 278 limbs were excluded EVT-first strategy. In the 138 EVT-first limbs, the target
from further observation by death. Of the 278 limbs, 39 lesions were FP in 55 cases, BK bypass in 41 cases, and
were lost to major amputation. Of the 39 limbs that both in 42 cases. Three of 138 EVT procedures (1.7%) failed
needed major amputation, the target arteries were within 30 days; one surgical conversion and one repeated
occluded in nine limbs. In the rest of 30 limbs with pat- EVT was performed. Of the 138 limbs, 74 (53.6%) required
ent target arteries, major amputation was performed TLRs owing to restenosis or hemodynamically unsuc-
because of massive necrosis or uncontrollable infection. cessful primary revascularization, including 35 surgical
Journal of Vascular Surgery
Volume -, Number -
Table V. Multivariate proportional hazards analysis of risk factors against wound healinga
Factor RR (univariate)
EVT first 2.1997
Diabetes 1.1932
History of CAD 1.1897
Dialysis dependent 1.4758
EF < 40% 1.6995
Wheel chair bound 1.3954
Infection 1.3961
WIfI wound grade $ 2 1.4946
Heel wound 1.1300
CAD, Coronary artery disease; EF, ejection fraction; EVT, endovascular therapy; RR, risk ratio; SUR, surgical revascularization; WIfI, Wound, Ischemia, and
foot Infection.
a
Significant risk factors against wound healing were an EVT-first strategy (RR, 2.47), CHF (RR, 2.05), and WIfI wound grade (RR, 1.59).
b
Indicates statistical significance (P <. 05).
conversions and 39 repeat EVT procedures. Major am-
putations were performed on 26 limbs during the
observation period, and 83 of the 138 limbs (60.1%) ach-
ieved wound healing. Of the 83 limbs that achieved
wound healing, 10 had recurrent ischemic ulcer and
three were amputated owing to recurrent CLI. Of the 55
limbs that did not achieve wound healing, wound
observation was terminated by amputation in 23 and by
death in 31, and one limb still had unhealed ulcer at
3 years after primary revascularization (Table II).
Technical outcome. Primary patency at 1, 2, and 3 years
after revascularization was 66.1%, 59.1%, and 55.2%,
respectively, in the SUR-first group and 46.8%, 44.3%,
and 40.8%, respectively, in the EVT-first group. Secondary
patency at 1, 2, and 3 years was 85.1%, 81.2%, and 79.8%,
respectively, in the SUR-first group and 77.1%, 73.1%, and
66.0%, respectively, in the EVT-first group (data not
shown).
Traditional clinical outcome. Overall survival at 1, 2, and
3 years after revascularization was 73.0%, 59.3%, and
51.4%, respectively. According to the multivariate propor-
tional hazards analysis, significant risk factors against
overall survival were dialysis dependence (RR, 2.72) and
being wheel chair bound (RR, 2.09). The limb salvage
rate at 1, 2, and 3 years after revascularization was 86.9%,
84.4%, and 83.9%, respectively. Significant risk factors for
limb loss were the EVT-first strategy (RR, 2.44), dialysis
dependence (RR, 3.62), and WIfI grade (RR, 3.55). AFS at 1,
2, and 3 years after revascularization was 64.4%, 53.2%,
and 45.5%, respectively. Significant risk factors against
AFS were dialysis dependence (RR, 2.31), CHF (RR, 1.94),
and being wheel chair bound (RR, 2.36; data not shown).
Analysis of wound healing. The cumulative proportion
of limbs that achieved wound healing at 3 years after pri-
mary revascularization with or without each risk factor
are shown in Table IV. Significant risk factors against
Okazaki et al 5
P RR (multivariate) P
<.0001b 2.4728 <.0001b
.2597 d d
.2328 d d
.0070b 1.2176 0.2039
.0194b 2.0543 .0015b
.0452b 1.0832 .6558
b
.0223 1.2567 .2056
.0078b 1.5934 .0086b
.6253
wound healing rate included EVT-first strategy, dialysis
dependence, CHF, being wheel chair bound, and WIfI
wound grade of $2.
The proportion of healed wounds 1, 2, and 3 years after
primary revascularization was 64.0%, 69.7%, 70.5%,
respectively (Fig 1). According to the multivariate Cox
regression model, significant risk factors against wound
healing were EVT-first strategy (RR, 2.47), CHF (RR, 2.05),
and WIfI grade (RR, 1.59; Table V). The mean WHT in
the subgroup of limbs that achieved wound healing
(n ¼ 196) was 143.7 days. An EVT-first strategy and wound
infection contributed to significantly longer WHT
(Table VI).
We analyzed ischemic wound recurrence in the sub-
group of limbs that achieved wound healing and found
that the proportion of limbs free from CLI recurrence 1,
2, and 3 years after wound healing was 91.7%, 86.6%,
83.1%, respectively. We found no significant risk factor
for CLI recurrence (data not shown). WFP was also
analyzed in the subgroup of limbs that achieved wound
healing (n ¼ 196), and the mean was 711.0 days. Among
analyzed risk factors, an EVT-first strategy, history of coro-
nary artery disease, and dialysis dependence were associ-
ated with significantly shorter WFPs (Table VII).
WRAFS at 1 and 2 years after achievement of wound
healing were 76.9% and 64.2% (Fig 2). Multivariate pro-
portional hazards analysis of risk factors against WRAFS
are shown in Table VIII. Significant risk factors against
WRAFS were the history of coronary artery disease (RR,
1.68), dialysis dependence (RR, 2.03), and being wheel
chair bound (RR, 1.64; Table VIII).
DISCUSSION
Historically, the outcomes of patients with peripheral
arterial disease have been evaluated primarily with
technical parameters such as graft patency or TLR. In
patients with CLI, clinical limb outcome was considered
6 Okazaki et al Journal of Vascular Surgery
--- 2017

Table VI. Mean wound healing time (WHT) in the sub- Table VII. Mean wound-free period (WFP) in the subgroup
group of limbs that achieved wound healing within 3 years of limbs that achieved wound healing within 3 years after
after primary revascularizationa primary revascularizationa
Factor No. WHT P Factor No. WFP P
Limbs achieved wound healing 196 143.7 Limbs achieved wound healing 196 711.0
EVT-first or SUR-first strategy EVT-first or SUR-first strategy
EVT 83 194.4 <.0001 b
EVT 83 639.7 .0127b
SUR 113 106.4 SUR 113 763.4
Diabetes mellitus Diabetes mellitus
Yes 135 148.4 .5097 Yes 135 721.5 .5263
No 61 133.3 No 61 687.7
History of CAD History of CAD
Yes 108 149.5 .5457 Yes 108 664.5 .0358b
No 88 136.6 No 88 768.2
Dialysis-dependent ESRD Dialysis-dependent ESRD
Yes 97 154.5 .3124 Yes 97 612.5 <.0001b
No 99 133.1 No 99 807.6
CHF (LVEF < 40%) CHF (LVEF < 40%)
Yes 19 183.0 .2484 Yes 19 594.5 .1007
No 170 137.4 No 170 730.4
Wheel chair bound Wheel chair bound
Yes 45 129.5 .4637 Yes 45 644.7 .1419
No 151 147.9 No 151 730.8
WIfI wound grade WIfI wound grade
1 73 128.8 .2782 1 73 714.4 .9156
$2 123 152.5 $2 123 709.0
Wound infection Wound infection
b
Yes 77 175.0 .0165 Yes 77 684.8 .3933
No 119 123.4 No 119 728.1
Heel wound Heel wound
Yes 17 139.4 .9008 Yes 17 624.2 .2781
No 179 144.1 No 179 719.3
CAD, Coronary artery disease; CHF, congestive heart failure; ESRD, end- CAD, Coronary artery disease; CHF, congestive heart failure; ESRD, end-
stage renal disease; EVT, endovascular therapy; LVEF, left ventricular stage renal disease; EVT, endovascular therapy; LVEF, left ventricular
ejection fraction; SUR, surgical revascularization; WIfI, wound, ischemia, ejection fraction; SUR, surgical revascularization; WIfI, wound, ischemia,
and foot infection. and foot infection.
a a
An EVT-first strategy and wound infection contributed to significantly An EVT-first strategy, history of CAD, and dialysis dependence
longer WHT. contributed to significantly shorter WFP.
b
Indicates statistical significance (P < .05). b
Indicates statistical significance (P < .05).

successful when the limb was rescued from major
amputation. However, the “limb salvage rate” does not
always represent successful limb outcome, because a
significant number of patients die before their symptoms
are relieved.
Most recent studies have used AFS as a clinical
endpoint.9,10 AFS is an excellent endpoint for evaluating
overall clinical outcome; however, it does not indepen-
dently assess limb outcome and survival. The uncertainty
about limb salvage in the subgroup of patients who die
before symptom relief (wound healing or pain relief)
makes the analysis of limb outcome difficult. Attempts
to determine the independent factors that affect
limb salvage have failed because these factors are
overshadowed by survival factors. Moreover, AFS also
does not always indicate successful limb outcomes,
because patients may survive a long time without major
amputation but with painful ischemic wounds.
The achievement of wound healing is a clear-cut
indicator for evaluating the outcome of limbs with
ischemic wounds.11 After the establishment of
“completely healed foot lesion” as an improved criteria
by Rutherford et al12 in 1997, investigators began to
analyze wound healing and its predictors. The reported
mean WHT varies from 47 to 264 days depending on the
clinical background. The suggested risk factors for failed
or delayed wound healing are diabetes.6 low serum al-
bumin level, ESRD, wound location on the heel,
Journal of Vascular Surgery
Volume -, Number -
n=196, mean WFP=773.0 days
1 year 2 years
WRAFS (%) 76.9 64.2%
Limbs at risk 150 118
Fig 2. Kaplan-Meier analysis of wound recurrence and
amputation-free survival (WRAFS). WFP, Wound-free
period.
Rutherford wound class 6,7 nonambulatory status, and
the targeting of indirect angiosomes during revasculari-
zation.5,13,14 These factors involved in delayed wound
healing can be classified into three groups: circulatory,
wound related, and systemic. Circulatory factors may
include the degree of peripheral arterial disease lesions,
type of revascularization, peripheral runoff, graft/EVT
patency, or targeting of angiosomes. Wound-related
factors include wound size, depth, location,15 infection,
and care. Together, circulatory and wound factors may
directly influence limb outcome, and they are summa-
rized and incorporated into the WIfI staging system.16
Systemic factors such as diabetes, poor nutrition, CHF,
ESRD, and nonambulatory status reportedly influence
limb outcome evaluated according to AFS. However,
these systemic risk factors may not directly influence
limb outcome, but may affect AFS because they have
life-limiting impacts.
The same situation applies when the clinical end point
is wound healing. If the general condition of patients is
poor, they may be unable to survive long enough to
achieve wound healing. Accordingly, a dilemma arises
in analyzing limbs that can no longer be observed owing
to death or amputation before wound healing. Simple
yes/no results about whether a limb achieved wound
healing are influenced by life-limiting factors.
A continuous scale analysis of WHT, similar to Kaplan-
Meyer analysis (wound healing rate), may be possible if
limbs that do not achieve wound healing are interpreted
to have an infinite WHT, but the results are still
Okazaki et al 7
influenced by life-limiting factors because patients
must survive long enough to achieve wound healing.
Therefore, these methods are effective for evaluating
overall outcome but not robust for evaluating limb
outcome independently. In analyses of limb outcome
alone, the uncertainty about wound healing can be
excluded by evaluating WHT only in limbs that achieve
wound healing, thereby excluding limbs that do not.
This method is effective for analyzing wound or limb
outcome independently but, inevitably, the data are
severely biased in this population, which does not reflect
clinical reality. Previous studies have reported that ESRD
is associated with a lower wound healing rate. However,
a recent study of WHT showed that ESRD does not signif-
icantly prolong WHT.13 Similarly, our analysis of risk fac-
tors for longer WHT seems to have extracted wound
healing factors alone because the results excluded sys-
temic factors, such as ESRD.
Even when limbs achieve wound healing, there is a
question about how long they can maintain a wound-
free state. Therefore, we introduced the parameters
WFP and WRAFS. These concepts are new and may be
influenced by the recurrence of ischemic wounds owing
to the loss of bypass grafts or EVT site patency, amputa-
tion, or death. WFP is the period during which patients
maintain a wound-free state. To achieve a longer WFP,
the wound must heal quickly and the patient must sur-
vive longer. WFP and WRAFS can be used as a compre-
hensive prognostic indicator similar to recurrence-free
survival in malignant disease. Longer WFP represent pos-
itive wound, limb, and life outcome.
We selected these multiple endpoints and adjusted our
limb subgroups depending on which outcome we were
evaluating. Our results showed that a blood circulation
factor (type of revascularization) and a wound factor
(WIfI wound grade) affected the achievement of wound
healing and WHT directly and independently, whereas
systemic factors (ESRD, CHF, and ambulatory status)
affected the achievement of wound healing and the
maintenance of a wound-free state indirectly by limiting
survival.
Several limitations of our study must be acknowledged.
This study was a retrospective, single-center study and
selection of revascularization procedure was not ran-
domized. Patient background was not controlled and
the population contained a higher percentage of
dialysis-dependent patients compared with Western
countries. Wound location, revascularization procedure,
target artery, and wound management were not strati-
fied. WIfI wound grade was determined via retrospective
review of photographs or descriptions of wounds in the
medical records. WHT may be overestimated because
wound status was not checked daily in outpatients
and, depending on the frequency of clinic visits, may
have introduced error of up to approximately 1 month.
8 Okazaki et al Journal of Vascular Surgery
--- 2017

Table VIII. Multivariate proportional hazards analysis of risk factors against wound recurrence and amputation-free survival
(WRAFS)a
Factor RR (univariate) P RR (multivariate) P
EVT first 1.5168 .0420 b
1.3634 .1414
Diabetes 0.8640 .4959 d d
History of CAD 1.7829 .0050b 1.6873 .0155b
Dialysis dependent 2.2624 <.0001b 2.0378 .0006b
EF < 40% 1.5179 .2188 d d
Wheel chair bound 1.5934 .0418b 1.6412 .0394b
Infection 1.0304 .8857 d d
WIfI wound grade $ 2 1.1199 .5878 d d
Heel wound 1.5864 .1751 d d
CAD, Coronary artery disease; CHF, congestive heart failure; EF, ejection fraction; ESRD, end-stage renal disease; EVT, endovascular therapy; LVEF, left
ventricular ejection fraction; RR, risk ratio; WIfI, wound, ischemia, and foot infection.
a
Significant risk factors were the history of CAD (RR, 1.68), dialysis dependence (RR, 2.03), and being wheel chair bound (RR, 1.64).
b
Indicates statistical significance (P < .05).

CONCLUSIONS critical limb ischemia? A critical reappraisal. Eur J Vasc

In this study, EVT was associated with longer WHT,
reduced wound healing rate, and shorter WFP compared
with SUR. WIfI grade was associated with longer WHT
and reduced wound healing rate, but not associated
with shorter WFP. Systemic conditions such as dialysis
dependence, CHF, and being wheel chair bound were
associated with a reduced wound healing rate and
shorter WFP, presumably because they limited life
expectancy. WHT and WFP are useful criteria for evalu-
ating limb outcomes in CLI patients.
AUTHOR CONTRIBUTIONS
Conception and design: JO
Analysis and interpretation: JO, YM
Data collection: JO, DM, KT, MI, SK, KM, TF
Writing the article: JO
Critical revision of the article: DM, KT, MI, SK, KM, TF, YM
Final approval of the article: JO, DM, KT, MI, SK, KM, TF, YM
Statistical analysis: JO, TF
Obtained funding: JO, DM, KT, MI, SK, KM, TF
Overall responsibility: JO
REFERENCES
1. Bradbury AW, Ruckley CV, Fowkes FG, Forbes JF, Gillespie I,
Adam DJ. Bypass versus angioplasty in severe ischemia of
the leg (BASIL): multicenter, randomized controlled trial.
Lancet 2005;366:1925-34.
2. Schanzer A, Mega J, Meadows J, Samson RH, Bandyk DF,
Conte MS. Risk stratification in critical limb ischemia: Deri-
vation and validation of a model to predict amputation-free
survival using multicenter surgical outcomes data. J Vasc
Surg 2008;48:1464-71.
3. Iida O, Soga Y, Yamaguchi Y, Hirano K, Kawasaki D,
Yamaoka T, et al. Clinical efficacy of endovascular ther-
apy for patients with critical limb ischemia attributable
to pure isolated infrapopliteal lesions. J Vasc Surg
2013;57:974-81.
4. Hoffmann U, Schulte KL, Heidrich H, Rieger H, Schellong S.
Complete ulcer healing as primary endpoint in studies on
Endovasc Surg 2007;33:311-6.
5. Lejay A, Georg Y, Tartaglia E, Gaertner S, Geny B, Thaveau F,
et al. Long-term outcomes of direct and indirect below-the
knee open revascularization based on the angiosome
concept in diabetic patients with critical limb ischemia. Ann
Vasc Surg 2014;28:983-9.
6. Soderstorm M, Arvela E, Alback A, Aho PS, Lepantalo M.
Healing of ischemic tissue lesions after infrainguinal bypass
surgery for critical leg ischemia. Eur J Vasc Surg 2008;36:
90-5.
7. Azuma N, Uchida H, Kokubo T, Koya A, Akasaka N,
Sasajima T. Factors influencing wound healing of critical
limb ischaemic foot after bypass surgery: is the angiosome
important in selecting bypass target artery? Eur J Vasc
Endovasc Surg 2012;43:322-8.
8. Chung J, Bartelson BB, Hiatt WR, Peyton BD, McLafferty RB,
Hopley CW, et al. Wound healing and functional outcomes
after infrainguinal bypass with reversed saphenous vein for
critical limb ischemia. J Vasc Surg 2006;43:1183-90.
9. Conte MS, Geraghty PJ, Bradbury AW, Hevelone ND,
Lipsitz SR, Moneta GL, et al. Suggested objective perfor-
mance goals and clinical trial design for evaluating catheter-
based treatment of critical limb ischemia. J Vasc Surg
2009;50:1462-73.
10. Conte MS. Understanding objective performance goals for
critical limb ischemia trials. Semin Vasc Surg 2010;23:129-37.
11. Goshima KR, Mills JL, Hughes JD. A new look at outcomes
after infrainguinal surgery: traditional reporting standards
systematically underestimate the expenditure of effort
required to attain limb salvage. J Vasc Surg 2004;39:330-5.
12. Rutherford RB, Baker JD, Ernst C, Jhonston KW, Porter JM,
Ahn S, et al. Recommended standards for reports dealing
with lower extremity ischemia: revised version. J Vasc Surg
1997;26:517-38.
13. Shiraki T, Iida O, Takahara M, Soga Y, Yamauchi Y, Hirano K,
et al. Predictors of delayed wound healing after endovascular
therapy of isolated infrapopliteal lesions underlying critical
limb ischemia in patients with high prevalence of diabetes
and hemodialysis. Eur J Vasc Endovasc Surg 2015;49:565-73.
14. Kret MR, Cheng D, Azarbal AF, Mitchell E, Liem TK,
Moneta GL, et al. Utility of direct angiosome revasculariza-
tion and runoff scores in predicting outcomes in patients
undergoing revascularization for critical limb ischemia.
J Vasc Surg 2014;59:121-8.
Journal of Vascular Surgery
Volume -, Number -
15. Soderstorm M, Aho PS, Lepantalo M, Alback A. The influence
of the characteristics of ischemic tissue lesions on ulcer
healing time after infrainguinal bypass for critical leg
ischemia. J Vasc Surg 2009;49:932-7.
16. Mills JL, Conte MS, Armstrong DG, Pomposelli FB,
Schanzer A, Sidawy AN, et al. The society of vascular surgery
Okazaki et al 9
lower extremity threatened limb classification system: risk
stratification based on wound, ischemia, and foot infection
(WIfI). J Vasc Surg 2014;59:220-334.
Submitted Feb 22, 2017; accepted Jul 16, 2017.