International Journal of Medical Physics, Clinical Engineering and Radiation Oncology, 2013, 2, 39-43
Published Online February 2013 (http://www.scirp.org/journal/ijmpcero)
http://dx.doi.org/10.4236/ijmpcero.2013.21006
Copyright © 2013 SciRes. IJMPCERO
Endovascular Treatment Options of Acute Limb Ischemia
Di Zhang, Wensheng Lou, Guoping Chen, Xindao Yin, Jianping Gu
Department of Radiation Physics, Nanjing Hospital, Nanjing Medical University, Nanjing, China
Email: yonzhang@mdanderson.org, songgao@mdanderson.org
Received October 14, 2012; revised November 15, 2012; accepted November 22, 2012
ABSTRACT
Acute limb ischemia is an urgent condition which occurs when there is an abrupt interruption of blood flow into an ex-
tremity usually because of either embolic or thrombotic vascular occlusion. Restoration of perfusion through early in-
tervention can decrease amputation and mortality. Contemporary treatment includes both surgery and endovascular te-
chniques. There is a rapid progress in endovascular intervention therapy. This article aims to make a comprehensive
review of the endovascular intervention options of acute limb ischemia.
Keywords: Acute Limb Ischemia; Arterial Embolus; Arterial Thrombosis; Endovascular Therapy
1. Introduction
Acute limb ischemia (ALI) of the lower extremities re-
mains a challenging clinical dilemma, which occurs when
there is an abrupt interruption of blood flow into an ex-
tremity usually because of either embolic or thrombotic
vascular occlusion [1,2]. The major clinical sign of ALI
is “6P”, i.e., pain, parasthesia, paralysis, pulselessness,
pallor and poikilothermia [3]. The overall prognosis is
poor, the mortality rate was 5.3% - 12%, amputation rate
was 5.3% - 14% [1,4-6]. When profound ischemia ensues,
this represents an emergency in which restoration of per-
fusion through early intervention can lead to limb sal-
vage, whereas delay may result in significant morbidity,
including limb loss and, potentially, death. Therefore,
timely and correctly treatment of this disease is impor-
tant.
There are three management options in acute lower
limb ischemia: 1) clot removal by catheter-directed thro-
mbolysis with or without percutaneous mechanical thro-
mbectomy; 2) surgical thromboembolectomy followed
by correction of underlying arterial lesions; and 3) anti-
coagulation with continued observation. We can catego-
rize the level of the patient’s limb ischemia utilizing cli-
nical assessment of motor and sensory function and in-
terrogating ankle arterial flow velocity signals into seve-
ral levels [7]. As shown in Table 1, those with level I
ischemia, particularly if they have significant comorbi-
dities, can and should be treated with heparin and ob-
servation, being watched closely while attention is given
to treating associated comorbidities. Conversely, in pa-
tients with the same level of ischemia (I), i.e., active pa-
tients without significant comorbidities, it is practical to
proceed directly with endovascular revascularization
(CDT and, possibly, percutaneous mechanical thrombec-
tomy). This same management choice is equally appro-
priate for those at level IIA, and at both of these levels of
ischemia there should be sufficient time for restoration of
patency using either endovascular or open surgical tech-
niques [8]. In the past, the decisional breakpoint in
choosing between endovascular and surgical revascu-
larization came between class IIA and class IIB, Treat-
ment of ALI has shifted toward endovascular therapies
because of rapidly improving technology and delivery
systems unless there is a contraindication, such as pro-
found critical limb ischemia, renal dysfunction, or con-
trast allergy. Regardless of which kind of therapy to be
chosen, the patient should be immediately anticoagu-
lated to prevent further clot extension. This article aims
to make a comprehensive review of the endovascular op-
tions with intra-arterial thrombolysis and/or adjuvant en-
dovascular techniques.
2. Catheter Directed Thrombolysis (CDT)
The development of effective medications to dissolve oc-
cluding thrombus led to the increasing use of throm-
bolytic therapy in patients with ALI. Randomized control
trials have provided a rationale for thrombolysis as a first
step in patients with ALI vs immediate operative revas-
cularization. Catheter directed thrombolysis offers seve-
ral potential advantages to lessen the pitfalls of open
techniques. By utilizing an endovascular approach and
local anesthesia, the risks of general anesthesia are mini-
mized within a subgroup of patients that are at their
physiologic limits [9]. In addition, enzymatic dissolution
of thrombus allows for more effective clot resolution
particularly within distal arterial beds that are often re-
D. ZHANG ET AL.
40
Table 1. Clinical categories of acute limb ischemia.
Category Prognosis Sensory Loss Motor Deficit Arterial
Doppler Venous
Doppler
I: Viable No immediate threat None None AudibleAudible
IIA: Marginally threatened Salvageable if promptly treated Minimal (toes) or none None InaudibleAudible
IIB: Immediately threatened Salvageable if immediately
revascularized More than toes, rest pain Mild/Moderate InaudibleAudible
III: Irreversible Major tissue loss, permanent nerve
damage inevitable
Profound,
anesthetic
Profound,
paralysis Inaudible Inaudible
sistant to open thrombectomy.
There are many evidences from randomized trials com-
paring catheter-directed thrombolysis with operative re-
vascularization. Ouriel et al. published a study that has
come to be known as the Rochester trial. The investiga-
tors randomized 114 patients with acute limb ischemia of
fewer than 7 days duration to urokinase or surgical in-
tervention. At 1 year, the cumulative risk of amputation
(18%) was identical in the two groups, while the cumula-
tive survival rate was significantly improved in patients
randomized to the thrombolysis group (84% vs 58% at
12 months, p = 0.01). The mortality differences seemed
to be primarily attributable to an increased frequency of
in-hospital cardiopulmonary complications in the opera-
tive treatment group (49% vs 16%, p = 0.001). The bene-
fits of thrombolysis were achieved without significant
differences in the duration of hospitalization (median 11
days) and with only modest increases in hospital cost in
the thrombolytic treatment arm (median $15,672 vs
$12,253, p = 0.02). Thrombolysis was equally effective
in those with embolic and thrombotic occlusions, al-
though the survival benefit was greater for patients with
embolic occlusions [10]. The Thrombolysis or Peripheral
Arterial Surgery (TOPAS) investigators randomized 213
patients with acute lower extremity ischemia secondary
to native arterial or bypass graft occlusion of fewer than
14 days duration to a variable dose of recombinant uro-
kinase (rUK) or surgery. Among patients treated with
rUK, surgical operations were avoided in 46% of patients
and the magnitude of such procedures was reduced in
50% of cases. Survival and amputation-free survival at
12 months were similar in the rUK and surgical groups.
Amputation-free survival was similar in the two groups.
There was a trend toward a higher amputation-free sur-
vival among those randomized to surgery and signifi-
cantly more bleeding in those randomized to rUK.
Among patients treated with rUK, thrombus resolution
and clinical outcome were somewhat better for acute
bypass graft thrombosis than for native arterial occlusion.
For thrombi longer than 30 cm, post-hoc analysis sug-
gested that 1-year amputation-free survival was better
following thrombolytic treatment, potentially due to lack
of suitable outflow for a surgical procedure. In contrast,
shorter occlusions fared better with surgery [11]. Al-
though The Surgery versus Thrombolysis for Ischemia of
the Lower Extremity (STILE) trial was prematurely ter-
minated by the Data and Safety Monitoring Board, sub-
sequent analysis, however, offered important insight. Pa-
tients presenting with acute ischemia (14 or fewer days)
and randomized to thrombolysis had significantly better
limb salvage (89% vs 70%) and amputation-free survival
[12].
It is very important to monitor coagulation parameters
carefully which aim to reduce risk hemorrhagic compli-
cations (estimated at 6% to 12.5%) during CDT proce-
dure [3,9-11]. Fibrinogen levels should also be checked
as a level < 100 mg/dL which reflects systemic fibrinoly-
sis and an increased risk of bleeding [12].
The severity of acute limb ischemia should be consi-
dered as the basis for therapic decision if we could use
CDT. CDT has long re-flow time, which can aggravate
ischemic. ALI of a limb artery can occur in the presence
or absence of underlying atherosclerotic involvement. If
there is no underlying atherosclerotic disease of the limb,
the thromboembolic event usually presents with pro-
foundly ischemic lower extremity, whereas thromboem-
bolic occlusion of chronically diseased lower limb artery
may present only with mild progression of chronic sym-
ptoms because of the development of collateral vessels
[8]. If ALI is located in the chronically diseased artery,
CDT could be the first choice.
We should master contraindications for CDT in ALI in
case of fatal haemorrhage [2,13]. Absolute contraindica-
tions to thrombolysis are well-known and include active
bleeding, central nervous system injury, or major opera-
tive procedures within the preceding 2 weeks. Relative
contraindications include uncontrolled hypertension, re-
cent eye surgery, pregnancy, and intracranial neoplasms.
3. Percutaneous Mechanical Thrombectomy
(PMT)
CDT is associated with slow restoration of blood flow,
which may aggravate tissue damage. The advent of PMT
has allowed removing clot burden quickly with CDT bol-
stered by using of PMT devices and can be used in pa-
Copyright © 2013 SciRes. IJMPCERO
D. ZHANG ET AL. 41
tients with acute profoundly limb ischemia. The earlier
clot-removing methods are applied, the better the out-
comes. Although no randomized prospective study showed
that safety and efficacy of percutaneous mechanical thro-
mbectomy or combined with thrombolysis in the treat-
ment of ALI, some data has shown its advantage in fast-
reflow and improvement of patients’ symptoms, and
PMT may be the only available treatment option in pa-
tients at high risk for open surgery or with contraindica-
tions to pharmacologic thrombolysis. K. Kasirajan et al.
published that thrombus removal rate was 85%, using a
PMT catheter as an initial treatment for acute (<2 weeks)
and subacute (2 weeks to 4 months) arterial occlusion of
the limbs. Angiographic outcome was not dependent on
the duration of occlusion or the conduit type [14]. Ansel
GM and his colleagues showed results of treatment of
acute limb ischemia with a percutaneous mechanical thro-
mbectomy-based endovascular approach. Angiography
following PMT showed thrombus removal complete/
substantial 63.6%, partial 28.0%, and minimal 8.8%, res-
pectively, and amputation free survival of acute limb is-
chemia treated with PMT alone or in combination with
thrombolysis was 94.7% at five-year follow-up [15].
Oguzkurt L et al. reported that complete thrombus re-
moval with PMT was achieved in 90% with acute occlu-
sions. Amputation-free survival rate was 100% at one
month, 93% at one year, and 93% at two years [16].
There is a new effective method to dissolve occluding
thrombus reported recently which using isolated phar-
maco-mechanical thrombolysis-thrombectomy (IPMT) to
isolate the thrombus between two balloons and utilize
wire oscillation to increase the thombus-lytic exposure
surface area followed by aspiration. IPMT decreases sys-
temic lytic exposure, procedure time, and distal embo-
lism [17].
PMT has generally been used as an adjunctive method
to CDT, but reverse was also true, and CDT could be an
adjunctive to PMT. Total dose of thrombolytic drug used
was low because PMT removed most of the thrombi in
most cases [14,15]. Main disadvantages of PMT were the
need for larger vascular sheath for large guiding catheter
than usual and the dissection of the artery. It should be
noted that a patient’s symptoms may get transiently wor-
se as the thrombus fragments with distal emboli. Distal
embolisation has been reported in 10% of cases with the
PMT procedure [16]. So during PMT, the operator should
perform gently to reduce risk of distal embolisation and
vascular injury.
4. Angioplasty
Successful clot dissolution will unveil a “culprit” lesion
responsible for initiating thrombosis. Angioplasty is not
the preferred initial method of treatment for ALI, only in
patients considered unfit for thrombolysis or in patients
who had undergone unsuccessful surgical recanalisation
or treatment with residual stenosis, as well as dealing
with intraoperative complications remedies [18]. Arte-
riography following removal of the thrombus by throm-
bolysis, and/or PMT usually delineates the responsible
lesion and, a decision must then be made in regard to
how best to manage this lesion and eliminate the threat of
recurrence it poses. Discrete atherosclerotic lesions are
well managed by either balloon angioplasty or, if acces-
sible in the surgical field, by surgical revision with patch
angioplasty, both yielding durable results. However,
longer, more extensive lesions or multiple stenoses-in-
series are still best treated by bypass. Kashyap et al. re-
ported in an overwhelming majority of cases (91%), an
adjuvant procedure was performed to treat the “culprit
lesion” leading to thrombosis. This was a purely percu-
taneous endovascular procedure in 56%, an open surgical
procedure in 15% and a combined approach in 28% [4].
Plate et al. [19] reported that angioplasty ratio after CDT
and/or PMT was 68%; PTA was 60%, 5% stent implan-
tation, aneurysm exclusion 2%, and 1% endarterectomy.
Kim et al. reported 15 cases of ALI underwent stent im-
plantation, technical success rate was 100%. Duration of
the intervention ranged between 20 min and 50 min.
During the average follow-up of 14.4 months, no re-oc-
clusion or restenosis was observed by CT angiography or
duplex USG [18]. Oguzkurt et al. [16] reported using
long-duration balloon inflation with/ without stent place-
ment to fix artery dissection. The articles about stenting
in ALI gradual increase in recent years, with good clini-
cal efficacy and no occurrence of distal embolization [20,
21]. Stent-assisted recanalisation may be an effective
treatment for patients with acute limb ischaemia who are
considered unfit for thrombolysis or surgical recanalisa-
tion or who have severe ischaemia requiring immediate
recanalisation or occlusion in the iliac arteries. Berczi et
al. [22] treated seven acute thrombotic occlusions in the
iliac arteries by primary stent implantation without distal
embolisation.
5. Embolic Protection Devices (EPDs)
Distal embolization following percutaneous revasculari-
zation procedures is a universal phenomenon that has
been reported in various vascular beds, including the ca-
rotid and renal arteries and the lower extremities. Distal
embolization may make it necessitate the use of addi-
tional interventions, including thrombectomy or throm-
bolysis, resulting in longer procedure time, greater
amount of contrast used, and larger radiation exposure.
EPDs are considered as a ‘‘standard of care’’ during per-
cutaneous carotid interventions, coronary artery bypass
graft PCI, and even renal angioplasty based on recent da-
ta. Although the FDA has not ratified any EPDs for pe-
ripheral arterial, several works have shown the safety and
Copyright © 2013 SciRes. IJMPCERO
D. ZHANG ET AL.
42
feasibility of EPDs use in this vascular territory. There
was a report, the first time in 2003, about using EPDs in
peripheral vascular angioplasty, and filter membranes
were evaluated with scanning electron microscopy. Sys-
tem delivery was successful in all patients [23]. D. Siab-
lis et al. [24] investigated the feasibility, safety and distal
emboli protection capability during recanalization of
lower extremities’ acute and subacute occlusions. Tech-
nical success rate of deployment and utilization of the
filtration devices was 100%. Macroscopic particulate de-
bris was extracted from all the filters containing fresh
thrombus, calcification minerals, cholesterol and fibrin.
Shammas et al. [25] reported that there were no compli-
cations related to the release and recovery process of the
device. Although many researchers believe patients with
high risk for distal embolic, poor outflow tract and direct
plaque excision will derive the greatest benefit from the
use of this technology, large multicenter registry is need-
ed to define the exact effects for these devices, to im-
prove the technical design for this unique vascular bed,
and clearly to identify best indications [23-25].
6. Summary
The treatment methods chosen should be based on the
patient’s conditions and angiographic results during cli-
nical practice. Now more and more interventional thera-
pies for ALI which derive from combinations of several
procedures above are used to restore blood flow quickly
and reduce complications. CDT combined with PMT
plus PTA/stent for ALI can be effective and minimally
invasive. In the future, one can expect to improve endo-
vascular techniques to the point where it will more sig-
nificantly impact on the management of acute as well as
chronic lower extremity occlusive diseases.
REFERENCES
[1] J. L. ELiason, R. M. Wainess, M. C. Proctor, J. B. Dimi-
ck, J. A. Cowan Jr., et al., “A National and Single Institu-
tional Experience in the Contemporary Treatment of Acu-
te Lower Extremity Ischemia,” Annals of Surgery, Vol.
238, No. 3, 2003, pp. 382-390.
[2] C. Kyriakides, W. G. Austen Jr., Y. Wang, J. Favuzza, F.
D. Moore Jr., et al., “Neutrophil Mediated Remote Or-
gan Injury after Lower Torso Ischemia and Reperfusion Is
Selectin and Complement Dependent,” The Journal of
Trauma, Vol. 48, No. 1, 2000, pp. 32-38.
doi:10.1097/00005373-200001000-00006
[3] M. Dvorak, M. Vlasin, M. Dvorakova, P. Rauser and L.
Lexmaulova, “Heparin and Its Derivatives in the Treat-
ment of Arterial Thrombosis: A Review,” Veterinarni
Medicina, Vol. 55, No. 11, 2010, pp. 523-546.
[4] V. S. Kashyap, R. Gilani, J. F. Bena, M. Bannazadeh and
T. P. Sarac, “Endovascular Therapy for Acute Limb Is-
chemia,” Journal of Vascular Surgery, Vol. 53, No. 2, 2011,
pp. 340-346. doi:10.1016/j.jvs.2010.08.064
[5] Z. Mitrev, F. Beyersdorf, R. Hallmann, Y. Poloczek, K.
Ihnken, et al., “Reperfusion Injury in Skeletal Muscle:
Controlled Limb Reperfusion Reduces Local and Sys-
temic Complications after Prolonged Ischaemia,” Car-
diovascular Surgery, Vol. 2, No. 6, 1994, pp. 737-748.
[6] K. Kasirajan, F. P. Beavers, D. G. Clair, D. G. Clair, R.
Greenberg, et al., “Rheolytic Thrombectomy in the Mana-
gement of Acute and Subacute Limb Threatening Ische-
mia,” Journal of Vascular and Interventional Radiology,
Vol. 12, No. 4, 2001, pp. 413-420.
doi:10.1016/S1051-0443(07)61878-8
[7] R. B. Rutherford, J. D. Bake, C. Ernst, et al., Recom-
mended Standards for Reports Dealing with Lower Ex-
tremity Ischemia: Revised Version,” Journal of Vascular
Surgery, Vol. 26, No. 3, 1997, pp. 517-538.
doi:10.1016/S0741-5214(97)70045-4
[8] R. B. Rutherford, “Clinical Staging of Acute Limb Ische-
mia as the Basis for Choice of Revascularization Method:
When and How to Intervene,” Seminars in Vascular Sur-
gery, Vol. 22, No. 1, 2009, pp. 5-9.
doi:10.1053/j.semvascsurg.2008.12.003
[9] K. Ouriel and F. J. Veith, “Acute Lower Limb Isehemia:
Determinants of Outcome,” Surgery, Vol. 124, No. 4,
1998, pp. 336-342. doi:10.1016/S0039-6060(98)70139-4
[10] K. Ouriel, C. K. Shortell, J. A. DeWeese, et al., “A com-
parison of Thrombolytic Therapy with Operative Revas-
cularization in the Initial Treatment of Acute Peripheral
Arterial Ischemia,” Journal of Vascular Surgery, Vol. 19,
No. 6, 1994, pp. 1021-1030.
doi:10.1016/S0741-5214(94)70214-4
[11] K. Ouriel, F. J. Veith and A. A. Sasahara, “A Comparison
of Recombinant Urokinase with Vascular Surgery as Ini-
tial Treatment for Acute Arterial Occlusion of the Legs.
Thrombolysis or Peripheral Arterial Surgery (TOPAS)
Investigators,” The New England Journal of Medicine
Vol. 338, No. 16, 1998, pp. 1105-1111.
doi:10.1056/NEJM199804163381603
[12] The STILE Investigators, “Results of a Prospective Ran-
domized Trial Evaluating Surgery versus Thrombolysis
for Ischemia of the Lower Extremity. The STILE Trial:
The STILE Investigators (Appendix A),” Annals of Sur-
gery, Vol. 220, No. 3, 1994, pp. 251-268.
doi:10.1097/00000658-199409000-00003
[13] A. S. Ward, S. K. Andaz and S. Bygrave, “Thrombolysis
with Tissue—Plasminogen Activator: Results with a High-
Dose Transthrombus Technique,” Journal of Vascular S ur-
gery, Vol. 19, No. 3, 1994, pp. 503-508.
doi:10.1016/S0741-5214(94)70078-8
[14] K. Kasirajan, F. P. Beavers, D. G. Clair, R. Greenberg, et
al., “Rheolytic Thrombectomy in the Management of
Acute and Subacute Limb Threatening Ischemia,” Jour-
nal of Vascular and Interventional Radiology, Vol. 12,
No. 4, 2001, pp. 413-420.
doi:10.1016/S1051-0443(07)61878-8
[15] G. M. Ansel, C. F. Botti and M. J. Silver, “Treatment of
Acute Limb Ischemia with a Percutaneous Mechanical
Thrombectomy-Based Endovascular Approach: 5-Year
Limb Salvage and Survival Results from a Single Center
Copyright © 2013 SciRes. IJMPCERO
D. ZHANG ET AL.
Copyright © 2013 SciRes. IJMPCERO
43
Series,” Catheterization and Cardiovascular Interven-
tions, Vol. 72, No. 3, 2008, pp. 325-330.
doi:10.1002/ccd.21641
[16] L. Oguzkurt, U. Ozkan, B. Gumus, I. Coşkun, N. Koca, et
al., “Percutaneous Aspiration Thrombectomy in the Treat-
ment of Lower Extremity Thromboembolic Occlusions,”
Diagnostic and Interventional Radiology Diagnostic, Vol.
16, No. 1, 2010, pp. 79-83.
[17] R. Gupta and T. A. Hennebry, “Percutaneous Isolated
Pharmaco-Mechanical Thrombolysis-Thrombectomy Sys-
tem for the Management of Acute Arterial Limb Ischemia:
30-Day Results from a Single-Center Experience,” Cathe-
terization and Cardiovascular Interventions, Vol. 80, No.
4, 2012, pp. 636-643. doi:10.1002/ccd.24283
[18] C. Kim, W. Jeon, T. Shin, D. Choi, J. Kim, et al., “Stent-
Assisted Recanalisation of Acute Occlusive Arteries in
Patients with Acute Limb Ischaemia,” European Journal
of Vascular and Endovascular Surgery, Vol. 39, No. 1,
2010, pp. 89-96. doi:10.1016/j.ejvs.2009.09.023
[19] G. Plate, I. Jansson, C. Forssell, P. Weber and S. Oreds-
son, “Thrombolysis for Acute Lower Limb Ischaemia—A
Prospective, Randomised, Multicentre Study Comparing
Two Strategies,” European Journal of Vascular and En-
dovascular Surgery, Vol. 31, No. 6, 2006, pp. 651-660.
doi:10.1016/j.ejvs.2005.11.017
[20] J. Raja, G. Munneke, R. Morgan and A. M. Belli, “Stent-
ing in Acute Lower Limb Arterial Occlusions,” Cardio-
vascular and Interventional Radiology, Vol. 31, No. S2,
2008, pp. S41-S44. doi:10.1007/s00270-007-9052-3
[21] S. Yilmaz, T. Sindel and E. Lüleci, “Primary Stenting of
Embolic Occlusions in Iliac Arteries,” Journal of Endo-
vascular Therapy, Vol. 10, No. 3, 2003, pp. 629-635.
doi:10.1583/1545-1550(2003)010<0629:PSOEOI>2.0.C
O;2
[22] V. Berczi, S. M. Thomas, D. R. Turner, J. R. Bottomley,
T. J. Cleveland, et al., “Stent Implantation for Acute Iliac
Artery Occlusions: Initial Experience,” Journal of Vas-
cular and Interventional Radiology , Vol. 17, No. 4, 2006,
pp. 645-649. doi:10.1097/01.RVI.0000203918.91835.73
[23] C. W. König, B. Pusich, G. Tepe, H. P. Wendel, U. Hahn,
et al., “Frequent Embolization in Peripheral Angioplasty:
Detection with an Embolism Protection Device (Angio-
guard) and Electron Microscopy,” Cardiovascular and
Interventional Radiology, Vol. 26, No. 4, 2003, pp. 334-
349. doi:10.1007/s00270-003-2656-3
[24] D. Siablis, D. Karnabatidis, K. Katsanos, P. Ravazoula, P.
Kraniotis, et al., “Outflow Protection Filters during Per-
cutaneous Recanalization of Lower Extremities’ Arteri-
alocclusions: A Pilot Study,” European Journal of Radi-
ology, Vol. 55, No. 2, 2005, pp. 243-249.
doi:10.1016/j.ejrad.2004.07.010
[25] N. W. Shammas, E. J. Dippel, D. Coiner, G. A. Shammas,
M. Jerin, et al., “Preventing Lower Extremity Distal Em-
bolization Using Embolic Filter Protection: Results of the
PROTECT Registry,” Journal of Endovascular Therapy,
Vol. 15, No. 3, 2008, pp. 270-276. doi:10.1583/08-2397.1