Keith R Poskitt - Chronic Ulceration of The Leg

You might also like

Download as pdf or txt
Download as pdf or txt
You are on page 1of 5

VASCULAR SURGERY e II

Chronic ulceration of the ulceration is usually due to multilevel occlusive peripheral arte-
rial disease causing critical leg ischaemia. Other causes of leg

leg ulceration include vasculitis and malignancy.

Epidemiology
Keith R Poskitt
Chronic leg ulceration is generally considered to be a disease of
Manj S Gohel the Western world, with an overall prevalence of ulceration of
around 0.5e1% in the adult population,2 increasing to over 3%
in patients over 80 years.3 The female preponderance is more
Abstract apparent in patients >60 years. Studies have suggested that the
The assessment and treatment of patients with chronic leg ulcers rep- development of venous ulcers is related to severity of varicose
resents an enormous clinical and financial burden to community and veins, or previous deep vein thrombosis (DVT). It is certainly the
secondary care services in the United Kingdom. Most leg ulcers are case that many patients with venous ulcers do have evidence of
due to chronic venous hypertension, although arterial compromise, venous disease (usually reflux, sometimes deep venous occlu-
malignancy and vasculitis are also recognized causes. The assess- sion). However, as many patients with varicose veins or previous
ment of patients with chronic leg ulceration should include a detailed DVT do not develop ulcers, these relationships are not fully un-
history and clinical examination, supported by relevant investigations. derstood. Ulcers due to arterial disease are more common in men
Ankle brachial pressure index measurement and non-invasive venous and are rare in patients <50 years.
(arterial) imaging using colour duplex are the principal investigations.
The mainstay of treatment for patients with chronic venous ulceration Pathology and pathogenesis
is multilayer compression bandaging, applied by trained staff working
within a specialist service bridging primary and secondary care set- Venous ulceration
tings. Endovenous treatment of superficial venous reflux may also Patients who develop venous ulcers have longstanding high pres-
play an important role in improving outcomes. Patients with arterial sure in the veins of the leg.4 This ‘chronic venous hypertension’ is
compromise may require endovascular or surgical revascularization usually a result of incompetence (or reflux) in superficial or deep
to promote healing. In this article, the causes, assessment and man- veins due to faulty valves, but may also occur in patients with
agement of patients with chronic leg ulceration are discussed. venous occlusion or recanalization after DVT. Other factors such as
immobility, obesity, ankle stiffness, leg dependency and poor calf
Keywords Arterial ulcer; compression bandaging; superficial venous
muscle pump function may also contribute to venous hypertension.
surgery; vascular surgery; venous ulcer
In the initial stages, patients with venous hypertension may be
asymptomatic, but over time the patient may develop varicose
veins, the skin may become inflamed (venous eczema), pigmented
Definition (haemosiderinosis) or thickened and scaly (lipodermatosclerosis)
and eventually break down resulting in ulceration. The Clinical,
Chronic leg ulceration may be defined as: Etiologic, Anatomic, Pathophysiological (CEAP) classification is
 a breach in the epithelial integrity of the skin used to describe the severity of venous disease5 (Table 1). Although
 occurring between the knee and malleoli the sequence of skin changes is well described, the precise patho-
 of greater than 6 weeks’ duration. genesis of ulceration due to venous hypertension is poorly under-
stood. Proposed theories include the fibrin cuff theory (observation
Aetiology of peri-capillary fibrin cuffs which may reduce local oxygenation),6
the white cell trapping hypothesis (trapped white blood cells may
The vast majority (>90%) of chronic leg ulcers have a vascular
become activated resulting in cytokine release and local tissue
aetiology. Chronic venous hypertension is thought to be the
damage)7 and the growth factor trapping theory (growth factors
primary cause of around 70% of leg ulcers and a significant
important for healing are inhibited by large molecules which have
contributory factor in a further 15%1 (Figure 1). In addition to
leaked out of capillaries due to venous hypertension).8 The typical
the primary aetiology, other factors may also delay wound
location for a venous ulcer is the medial gaiter area of the leg,
healing and should be considered and treated where possible.
although lateral leg and foot ulceration (rare) may occur. Other
These include dependent oedema (which is often poorly
venous skin changes are commonly seen with the ulcer (Figure 2).
controlled), medical comorbidities, poor nutrition and medica-
tions (including steroids or immunosuppressive drugs). Arterial
Arterial ulceration
Risk factors for occlusive peripheral vascular disease include age,
male sex, smoking, dyslipidaemia, diabetes and hypertension.
Multilevel arterial stenosis or occlusion may result in critical limb
Keith R Poskitt MB ChB MD FRCS is a Consultant Surgeon at ischaemia and tissue loss. Although arterial ulceration is often
Cheltenham General Hospital, Gloucestershire, UK. Conflicts of seen on the foot (or involving the toes), more proximal leg ulcers
interest: none. may also be seen (Figure 3). Patients with a combination of
Manj S Gohel MB ChB MD FRCS FEBVS is a Consultant Vascular arterial and venous disease present a unique challenge, as
Surgeon at Addenbrooke’s Hospital, Cambridge, UK. Conflicts of identification of the primary ulcer cause (and main therapeutic
interest: none. target) may be difficult. A combination of clinical, anatomical

SURGERY 34:4 178 Ó 2016 Published by Elsevier Ltd.


VASCULAR SURGERY e II

 the impact of the ulcer on the patient


Causes of chronic leg ulceration  patient expectations of treatment.
Other causes* A detailed history may provide key clues to the aetiology of the
10% ulcer and should provide strong clinical suspicions to be confirmed
Arterial by physical examination and appropriate investigations. A history
5%
of DVT or varicose veins may indicate chronic venous hyperten-
sion, whereas a history of smoking, diabetes, other risk factors or
Mixed arterial symptoms of peripheral vascular disease may be more suggestive
and venous
15% of an arterial component. Advanced patient age and, large ulcer
size and long ulcer chronicity have been shown to be independent
risk factors for delayed ulcer healing and may be useful prognostic
indicators.10 Moreover, a detailed knowledge of previous surgery,
Venous co-existing illnesses, medications and patient occupation may be
70%
important in planning treatment.
General clinical examination may reveal systemic illness or
*Vasculitis, malignancy, haematological disorders, rheumatoid arthritis, nutritional deficiencies contributing to poor wound healing. The
pressure/trauma, diabetes
ulcerated skin should be examined to identify the ulcer location,
size, edges, signs of granulation and surrounding skin. Evidence
Figure 1 of typical venous skin changes and a granulating ulcer in the
and haemodynamic assessments can be used to guide patient medial gaiter area would be consistent with a chronic venous
management. ulcer, whereas a painful punched-out ulcer in a limb with absent
pulses may indicate an arterial cause. In order to monitor ulcer
Malignancy progression and response to treatment, ulcer assessment should
Skin malignancies may present as chronic leg ulcers and should include details of ulcer morphology, precise location and size,
be considered in all ulcers with suspicious features or poor facilitated by use of wound tracings or photography.
healing.9 Basal cell (60%) and squamous cell carcinomas (40%)
account for the vast majority. Malignant transformation (squa- Investigations
mous cell carcinoma) of chronic venous ulcers is an occasional, Ankle brachial pressure index (ABPI): all patients with chronic leg
but rare, occurrence (Marjolin’s ulcer). ulceration should undergo ABPI assessment to identify underlying
arterial compromise (Figure 4). Significant arterial disease may be
Other causes excluded if ABPI >0.85, although falsely raised ankle pressures may
Vasculitis, diabetes, rheumatoid arthritis and other systemic be present due to calcification in the vessel wall. Patients with ABPI
conditions (such as sickle cell disease or sarcoidosis) can cause <0.85 should be considered to have arterial compromise and may
leg ulceration and may be difficult to distinguish from venous require further specialist investigation or treatment.
ulcers. Biopsy of the ulcer edge for histological examination will
usually provide accurate diagnosis of uncommon causes for leg Colour venous duplex scan: colour venous duplex uses a
ulceration. combination of Doppler and B-mode ultrasound to accurately
map superficial and deep venous patency and competence in the
Diagnosis ulcerated leg. Duplex is widely accepted as the investigation of
Clinical history and examination choice to identify superficial venous reflux potentially amenable
The specific aims of patient assessment are to ascertain: to surgery or endovenous intervention. Colour duplex scanning
 the aetiology of the ulceration can also identify post-thrombotic scarring and occlusions in iliac
 other contributing factors that may delay healing and femoral veins.

Ulcer biopsy: as basal cell or squamous cell carcinoma may be a


Clinical, Etiological, Anatomical and Pathophysiological feature in up to 1% or more of chronic leg ulcers, wounds with a
(CEAP) classification of chronic venous disease suspicious appearance or no evidence of healing after 3 months
CEAP clinical stage Description should be biopsied.9 Ulcer edge biopsy may also help identify
ulcers due to vasculitis or diabetes, which may be difficult to
C0 Absence of any signs of venous disease distinguish from venous or arterial ulcers.
C1 Reticular veins
C2 Truncal varicose veins Other investigations: tests of venous haemodynamic function
C3 Oedema are generally limited to research studies, but may be of benefit in
C4 Skin changes (pigmentation, some cases. Preoperative digital photoplethysmography (PPG)
lipodermatosclerosis) using a tourniquet has been shown to predict improvements after
C5 Healed ulceration surgery and may identify good candidates for superficial venous
C6 Open ulceration intervention in patients with mixed superficial and deep venous
reflux.11 Further arterial investigations such as colour duplex or
Table 1

SURGERY 34:4 179 Ó 2016 Published by Elsevier Ltd.


VASCULAR SURGERY e II

Venous ulcers
General measures: patients should be advised to elevate the
limb where possible and continue regular exercise. Measures to
stop smoking and improve nutrition may also be beneficial.

Compression therapy: multilayer graduated compression


bandaging applied by trained personnel (providing 40 mmHg of
pressure at the ankle and 17e20 mmHg at the upper calf) is
accepted as the mainstay of treatment for chronic venous ulcer-
ation.14 Numerous bandaging regimens including short stretch,
multilayer systems and double-layered compression stockings
have been advocated. Once healed, patients should be advised to
Figure 2 Typical venous leg ulcer with associated skin changes of wear elastic stockings (class II) to reduce the risk of recurrent
chronic venous hypertension.
ulceration.15 Clinicians should recognize that compliance with
compression is likely to be poor, particularly as compression
stockings may be very difficult to put on and take off.

The importance of treating superficial reflux: understanding of


the importance of treating superficial venous reflux for patients
with chronic venous ulcers has increased in recent years. The
ESCHAR venous ulcer trial demonstrated that traditional super-
ficial venous surgery offered as an adjunct to compression ther-
apy reduces venous ulcer recurrence.16,17 However, no
improvements in healing rates were seen. These findings were
substantiated by a Dutch randomized study in 2006.18 Despite
the unequivocal clinical benefit, traditional varicose vein surgery
is not a popular treatment option for elderly patients with leg
ulcers.
The development and widespread uptake of minimally invasive
endovenous interventions has revolutionized the management of
Figure 3 Leg ulceration in a patient with arterial compromise. patients with chronic venous disease. Novel endovenous treat-
ments such as laser ablation, radiofrequency ablation or foam
angiography may be necessary where ABPI <0.85 or inconclu- sclerotherapy have been shown to provide comparable or superior
sive (due to calcified vessels). Further specific investigations may technical success rates in the short and medium-term.19 Ultra-
be required in cases suspicious of vasculitis. There has been a sound guided foam sclerotherapy perhaps represents the ‘least’
recent growth of interest in deep venous occlusive disease, which invasive endovenous intervention and has gained support from
may be an important contributing factor in more patients with many experts as the treatment of choice in this population.20e22 As
venous ulcers than previously thought. Specific deep venous all of the endovenous ablation (either thermal or chemical) treat-
investigations such as magnetic resonance venography (MRV), ment modalities can be offered in an outpatient setting using only
computerized tomogram venography (CTV) or invasive venog- local (or no) anaesthesia, they are all likely to be more acceptable
raphy may be useful in some patients. then open venous surgery to elderly patients with venous leg ul-
cers. The clinical benefit of modern endovenous treatments is
being evaluated in the NIHR funded Early Venous Reflux Ablation
Management
(EVRA) trial,23 which is currently recruiting patients.
Patients with chronic leg ulcers are best managed by specialist,
multidisciplinary nurse-led teams with close links to primary and Treatment of deep venous disease: there has been a significant
secondary care services. With the regular development and increase in the popularity of deep venous stenting procedures in
introduction of new wound dressings and products, it is recent years.24 With the introduction of specialist stents, prom-
important to emphasize that treatments should be guided by ising technical success and patency rates have been reported.
evidence-based, nationally agreed protocols. A number of na- While the precise role of deep venous stenting in patients with
tional and international guidelines have been published to guide leg ulcers remains unclear, there are undoubtedly some patients
the assessment and management of patients with chronic venous with significant (usually post-thrombotic) deep venous occlusive
disease and ulceration.12,13 disease, who may benefit from these novel procedures.
The primary aims of treatment are to correct the underlying
cause of ulceration and address any exacerbating factors to Other treatments: wound debridement may be useful in specific
promote rapid ulcer healing and reduce the risk ulcer recurrence. cases and can be performed with dressings, larvae or surgery.
A multidisciplinary approach involving specialist nurses, Healing of larger ulcers may be accelerated using pinch or
vascular surgeons, rheumatologists, dieticians, dermatologists meshed split skin grafting. The routine use of systemic antibiotics
and other health professionals may be required. does not improve healing and antibiotic use should be limited to

SURGERY 34:4 180 Ó 2016 Published by Elsevier Ltd.


VASCULAR SURGERY e II

important component of leg ulcer management, even when


chronic venous hypertension is not the primary cause.

Challenges in managing chronic leg ulceration


Despite advances in our understanding of chronic leg ulceration,
the management of this patient group remains sub-optimal. A
number of studies have demonstrated the benefits of specialist
nurse-led community leg ulcer clinics with close links between
primary and secondary care services, but the model of leg ulcer
care in the United Kingdom remains inconsistent. The majority of
patients with leg ulcers are assessed and managed by community
nursing staff, usually with appropriate compression bandaging.
However, pathways for the assessment and treatment of super-
ficial venous reflux are poorly defined, meaning that patients
often do not receive intervention for superficial incompetence
despite the proven advantages. Despite a strong evidence base,
implementation of best practice for patients with chronic leg ul-
cers is frequently poor.
Unfortunately, the current NHS reimbursement and funding
structure may act as a disincentive against greater involvement
of secondary care services in the management of these patients.
Figure 4 Ankle brachial pressure index (ABPI) assessment. Referrals for secondary care are often reserved for patients that
fail to respond after many months (or years) of standard leg
patients with cellulitis. Evidence supporting use of systemic
ulcer treatment. While this may appear to be a logical approach,
medications or topical growth factors is lacking. Novel and
ulcer chronicity is a major predictor of poor ulcer healing.
exciting recent developments include ‘spray-on skin’ technol-
Therefore, early identification of patients with risk factors of
ogy25 (utilizing autologous dermal cells harvested and adminis-
poor ulcer healing (which are well described) and referral for
tered in a single outpatient visit) and dressings that can deliver
specialist secondary care (such as early treatment of superficial
topical oxygen. Although early reports have been encouraging,
reflux) would seem a more effective approach. There is a lack of
larger studies with long-term follow-up are eagerly awaited.
high profile national initiatives, guidelines or service frame-
Arterial ulcers works for patients with chronic leg ulceration, reinforcing the
Revascularization is usually indicated where arterial compromise widely held belief that this is an unglamorous condition
is the primary cause of ulceration. This may be performed by affecting a largely elderly, uncomplaining and inconspicuous
endovascular intervention (angioplasty or stent) or revasculari- population.27
zation surgery (endarterectomy or bypass). Management should
include aggressive optimization of cardiovascular risk factors Complications
and careful diabetic control. Any ulcer may become infected and cause local cellulitis or
systemic sepsis. Although some patients undoubtedly require
Mixed arterial/venous ulcers
prolonged treatment with antibiotics and may even require
The management of mixed arterial/venous leg ulcers is contro-
hospital admission, many patients are treated with antibiotics on
versial and should be determined on an individual basis. The use
the basis of a positive growth from a wound swab. Studies have
of modified multilayer compression bandaging (25e30 mmHg at
shown that most chronic ulcers have bacterial colonization and
ankle) has been shown to be safe for patients with ABPIs be-
empirical treatment is unlikely to improve outcomes. Other
tween 0.5 and 0.85 within a specialist leg ulcer service with close
complications include immobility (due to pain), reduced quality
vascular surgical support.26 However, patients should be advised
of life and loss of independence.
to remove bandaging immediately if severe leg pain develops and
intensive follow-up is essential in this high-risk group. Where Prognosis and explanation to patient
modified compression is not tolerated, revascularization may be
necessary to facilitate ulcer healing. Careful and regular assess- Venous ulcers
ment by a specialist leg ulcer service is particularly important in Healing: patients may be advised that around 65% of venous
these high-risk patients. ulcers treated with effective multilayer compression heal within
24 weeks, although reported healing rates range from 36% to
Other ulcers 83%. However, up to 20% of chronic venous ulcers may remain
Ulcers proven to be malignant on biopsy should be treated by unhealed after 1 year, despite compression bandaging. Recog-
surgical excision with adequate margins and may require skin nized risk factors for delayed healing include advanced patient
grafting to facilitate healing. For patients with ulcers secondary age, ulcer chronicity and ulcer size.10 Scoring systems to predict
to systemic illness (such as vasculitis or diabetes), treatment the probability of healing for individual patients have been
should aim to address the underlying condition. Adequate described,28 but are not in widespread use. The role of endove-
oedema control (possibly with compression therapy) is an nous interventions to improve ulcer healing is debatable,

SURGERY 34:4 181 Ó 2016 Published by Elsevier Ltd.


VASCULAR SURGERY e II

although results from the EVRA study may help resolve this 11 Gohel MS, Barwell JR, Heather BP, et al. The predictive value of
uncertainty. Level one evidence is lacking, but a number of haemodynamic assessment in chronic venous leg ulceration. Eur
prospective studies have suggested that 24-week ulcer healing J Vasc Endovasc Surg 2007 Jun; 33: 742e6.
rates may increase to around 80% with early treatment of su- 12 Wittens C, Davies AH, Bækgaard N, et al. Editor’s choice e
perficial reflux.22 management of chronic venous disease: clinical practice guide-
lines of the European Society for Vascular Surgery (ESVS). Eur J
Recurrence: 1-year venous ulcer recurrence rates as high as Vasc Endovasc Surg 2015; 678e737.
69% have been reported. However, within the ESCHAR trial, 13 O’Donnell TF, Passman MA. Clinical practice guidelines of the
recurrence rates for patients treated with compression and Society for Vascular Surgery (SVS) and the American Venous
venous surgery were 12% at 1 year and 31% at 4 years. These Forum (AVF)eManagement of venous leg ulcers. Introduction.
were significantly lower than recurrence rates for patients J Vasc Surg 2014 Aug; 60: 1Se2.
treated with compression alone (28% at 1 year and 56% at 4 14 O’Meara S, Cullum N, Nelson EA. Compression for venous leg
years).16,17 ulcers. Cochrane Database Syst Rev 2012. Issue 11. Art. No.:
CD000265.
Arterial ulcers 15 Nelson EA, Bell-Syer SEM. Compression for preventing recur-
In patients with ulceration due to severe arterial compromise, rence of venous ulcers. Cochrane Database Syst Rev 2014. Issue
healing rates are unlikely to exceed 50% despite an aggressive 9. Art. No.: CD002303.
policy of revascularization as these patients are often unfit or 16 Barwell JR, Davies CE, Deacon J, et al. Comparison of surgery
unsuitable for arterial intervention.26 and compression with compression alone in chronic venous
ulceration (ESCHAR study): randomised controlled trial. Lancet
Mixed arterial/venous ulcers 2004 Jun 5; 363: 1854e9.
In a recent observational study, 36 week healing rates of 68% 17 Gohel MS, Barwell JR, Taylor M, et al. Long term results of
were achieved with modified compression therapy for patients compression therapy alone versus compression plus surgery in
with mixed arterial/venous ulcers and revascularization was chronic venous ulceration (ESCHAR): randomised controlled trial.
performed in only 10% of patients.26 A BMJ 2007 Jul 14; 335: 83.
18 van Gent WB, Catarinella FS, Lam YL, et al. Conservative versus
surgical treatment of venous leg ulcers: 10-year follow up of a
REFERENCES randomized, multicenter trial. Phlebology 2015 Mar; 30: 35e41.
1 Ghauri ASK, Currie IC, Grabs AJ, Whyman MR, Farndon JR, 19 van den Bos R, Arends L, Kockaert M, Neumann M, Nijsten T.
Poskitt KR. Improving the diagnosis of chronic leg ulcers: a one- Endovenous therapies of lower extremity varicosities: a meta-
stop vascular assessment clinic in a community service. Phle- analysis. J Vasc Surg 2009 Jan; 49: 230e9.
bology: J Venous Dis 1998 Dec 1; 13: 148e52. 20 Grover G, Tanase A, Elstone A, Ashley S. Chronic venous leg
2 Nelze n O, Bergqvist D, Lindhagen A. The prevalence of chronic ulcers: effects of foam sclerotherapy on healing and recurrence.
lower-limb ulceration has been underestimated: results of a vali- Phlebology 2016; 31: 34e41.
dated population questionnaire. Br J Surg 1996 Feb; 83: 255e8. 21 Kulkarni SR, Slim FJA, Emerson LG, et al. Effect of foam sclero-
3 Cornwall JV, Dore CJ, Lewis JD. Leg ulcers: epidemiology and therapy on healing and long-term recurrence in chronic venous
aetiology. Br J Surg 1986 Jan; 73: 693e6. leg ulcers. Phlebology 2013 Apr; 28: 140e6.
4 Burnand KG, Clemenson G, Whimster I, Gaunt J, Browse NL. The 22 Pang KH, Bate GR, Darvall KAL, Adam DJ, Bradbury AW. Healing
effect of sustained venous hypertension on the skin capillaries of and recurrence rates following ultrasound-guided foam sclero-
the canine hind limb. Br J Surg 1982; 69: 41e4. therapy of superficial venous reflux in patients with chronic venous
5 Eklo€ f B, Rutherford RB, Bergan JJ, et al. Revision of the CEAP ulceration. Eur J Vasc Endovasc Surg 2010 Dec; 40: 790e5.
classification for chronic venous disorders: consensus statement. 23 EVRA (Early Venous Reflux Ablation) ulcer trial. Available from:
J Vasc Surg 2004; 40: 1248e52. www.evrastudy.org.
6 Burnand KG, Whimster I, Naidoo A, Browse NL. Pericapillary fibrin 24 Seager MJ, Busuttil A, Dharmarajah B, Davies AH. A systematic
in the ulcer-bearing skin of the leg: the cause of lip- review of endovenous stenting in chronic venous disease sec-
odermatosclerosis and venous ulceration. Br Med J (Clin Res Ed) ondary to iliac vein obstruction. Eur J Vasc Endovasc Surg 2016;
1982 Oct 16; 285: 1071e2. 51: 100e20.
7 Coleridge Smith PD, Thomas P, Scurr JH, Dormandy JA. Causes 25 Kirsner RS, Marston WA, Snyder RJ, Lee TD, Cargill DI, Slade HB.
of venous ulceration: a new hypothesis. Br Med J (Clin Res Ed) Spray-applied cell therapy with human allogeneic fibroblasts and
1988 Jun 18; 296: 1726e7. keratinocytes for the treatment of chronic venous leg ulcers: a
8 Falanga V, Eaglstein WH. The ‘trap’ hypothesis of venous ulcer- phase 2, multicentre, double-blind, randomised, placebo-
ation. Lancet 1993 Apr 17; 341: 1006e8. controlled trial. Lancet 2012 Sep 15; 380: 977e85.
9 SenetP, CombemaleP, DebureC, etal.Malignancyandchronicleg 26 Humphreys ML, Stewart A, Gohel MS. Management of mixed
ulcers:thevalueofsystematicwoundbiopsies:a prospective, multi- arterial and venous leg ulcers. Br J Surg 2007 Sep; 94: 1104e7.
center,cross-sectional study.ArchDermatol 2012Jun;148:704e8. 27 Bulbulia RA, Poskitt KR. The need for a National Service Frame-
10 Gohel MS, Taylor M, Earnshaw JJ, Heather BP, Poskitt KR, work for leg ulcers. Phlebology 2010 Oct; 25(suppl 1): 68e72.
Whyman MR. Risk factors for delayed healing and recurrence of 28 Kulkarni SR, Gohel MS, Wakely C, Minor J, Poskitt KR, Whyman MR.
chronic venous leg ulcersean analysis of 1324 legs. Eur J Vasc The ulcerated leg severity assessment score for prediction of venous
Endovasc Surg 2005 Jan; 29: 74e7. leg ulcer healing. Br J Surg 2007 Feb; 94: 189e93.

SURGERY 34:4 182 Ó 2016 Published by Elsevier Ltd.

You might also like