A. Prof. Anthony J. Dixon1, MBBS FACRRM FACSCM
Mary P. Dixon2, B Appl Sci (Nursing)
A. Prof. John B. Dixon3, MBBS FRACGP PhD
Method: Prospective study of 5950 skin lesions excised on 2394 patients.
Objective: To identify body sites at increased risk of post operative bleeding following skin cancer surgery procedures.
Results: Post operative bleeding incidence was 0.67% overall (40/5950). The incidence was increased at 2.16% when surgery was on or near the ear (9/416). (p<0.001) Bleeding incidence on the pinna was 2.24% (7/313) and the immediate post auricular incidence was 3.03% (1/33).
The only other body sites demonstrating a bleeding incidence over 1% were the nose (6/571), neck (3/199) and temple (2/137) regions. In each of these sites the increase was not significant compared to elsewhere on the body.
No bleeding complication was life threatening and most bleeds were managed with pressure and dressings. Only three patients required wound exploration to manage ongoing post operative bleeding.
Limitations: This study involved a single surgeon in Southern Australia and may not be indicative of broader skin surgical practice.
Conclusion: Postoperative bleeding following skin surgery is uncommon and is usually able to be managed conservatively. 1 in 2000 procedures might come to wound exploration due to ongoing bleeding. The ear, including immediate pre-auricular and post auricular sites are the only body regions demonstrating an increased incidence of post operative bleeding following skin cancer surgery. Attention to haemostasis during surgery and care with dressings following surgery may be more important when excisions are planned in and around the ear.
There is limited data to identify sites of skin that may be at higher risk of bleeding complications. It has been reported that surgery to the outer ear shares the bleeding risk with other parts of the body.216 Elsewhere, ear nose and throat surgery has been identified as at higher risk of unexpected readmission with bleeding complications cited as one such reason.217
Through a prospective study we identified incidence of post operative bleeding following skin surgery for all body sites to identify locations that may be at increased risk.
This study involved patients managed from 1 July 2002 to 28 February 2006 at the skin cancer referral centre, “Skincanceronly”, Geelong, Australia. Patients’ aspirin was continued and warfarin was not altered either before or following skin cancer surgery unless INR was over 3.0. In the event that post operative haemorrhage occurred, pressure, dressings and wound exploration were implemented as appropriate. This trial was approved by the Barwon Health Research and Ethics committee.
Surgical procedures included: modified16 margin control surgery12, 218; direct excision and closure of lesions; curettage; skin flaps; full thickness and partial thickness skin grafts; and wedge excision surgery.
Patients were excluded if they ceased their warfarin or aspirin prior to attendance or recorded a preoperative INR level over 3.0. Bipolar diathermy and ligatures were available for all cases. All full thickness wounds were closed with nylon or polyamide interrupted skin sutures. Absorbable deep sutures were used in closure only if sheath, cartilage, muscle or other layers were breached or if dermis and epidermal closures would benefit from separate closure.
Patients were followed up at least until removal of sutures. Expanded details of the methodology have been previously published.4
Demographic details were presented as percentage or mean ± standard deviation (SD) as appropriate. Analysis was univariate (Chi-square method) and multivariate; tested using binary logistic regression (forward and backward) and odds ratio beta-coefficients with 95% confidence intervals are shown. The SPSS 14.0.2 statistical software was used for all statistical analysis. A p-value of less than 0.05 was considered statistically significant.
A total of 5950 skin lesions from 2394 patients were treated by excision or curettage. There were 28 exclusions. Lesions managed included 3175 malignant lesions; 1436 SCCs (24.1%), 1381 BCCs (23.2%), 166 melanoma (2.8%), 24 lentigo maligna and 168 other cutaneous malignancies (2.8%).
Post operative bleeding
40 bleeds were recorded (0.67%). There were14 haematomas and 26 haemorrhages recorded.
There was one large bleed three weeks post surgery. The patient on warfarin had a therapeutic INR at the time of surgery but it rose to 7.4 in the weeks following surgery. He was the only person hospitalized due to post operative bleeding. Management involved ceasing warfarin and wound compression.
Three patients (two on warfarin) required wound exploration to control bleeding. One patient required vessel ligation and the other two patients had bleeding controlled with bipolar diathermy, Kaltostat ® and compression dressings. Two patients (neither on warfarin) with haematoma had their collection evacuated. All other patients had bleeding complications managed conservatively; pressure dressings and review.
Bleeding in relation to surgical technique
The lowest bleeding incidence was seen with wounds closed directly, (0.42%) and those managed by curettage, (0.20%)
Skin flap closure results in an increased bleeding risk, odds ratio (OR) 2.6, (1.4 – 4.9) P=0.02.
The types of skin flaps (n=1958) undertaken were; 488 transposition (24.9%), 341 bilobed (17.4%), 296 O to S (15.1%), 201 reducing opposed multilobed (ROM)6, 7 (10.3%), 103 A to T (5.3%), 100 V – Y (5.1%), 429 others (21.9%).
Skin grafts revealed a higher risk of bleeding of OR 8.5 (3.3 – 22) P<0.001.
Bleeding in relation to site on body
The bleeding incidence at different sites on the body is detailed in Table 1.
Procedures on and around the ear showed an increased incidence of bleeding complications, 2.24% (p<0.001) Ear bleeds were greater for all procedure types or whether taking warfarin or not, (Table 2).
Surgery on the ear was combined with surgery adjacent to the ear as frequently defects on the ear were repaired with skin grafts or flaps involving adjacent skin.
No other body site demonstrated a statistical increased incidence of post operative bleeding.
Table 1: The bleeding incidence following surgery to various body regions
% of all sites
Post operative bleeding
P value to rest *
Head and neck
Ear + pre & post auricular
Cheek & Zygoma
Chin & lips
Thigh and groin
* P values determined using chi square test comparison to other body locations
** Single bleeds preclude chi square analysis
Table 2: Comparison of procedure on and near the ear with procedures elsewhere.
Total ear procedures#
Total non ear procedures
Non ear bleeds
not on warfarin
Chi square analysis-**Chi square test not possible with only a single bleed.
# includes procedures on the ear as well as immediate pre and post auricular.
Table 3: The percentage of bleeding complication for patients grouped by the number of bleeding risk factors.
Overall bleeding rate
Percent of all cases managed
0 risk factors
1 risk factor
2 risk factors
3 risk factors
4 risk factors**
** Only 1 bleed experienced in patients with all 4 independent risk factors
Risk factors were age 67 or older, warfarin, ear location, and flap or graft closure
There is an increased risk of bleeding in patients having skin cancer surgery in and around the ear. Our analysis was multivariate and demonstrated that all procedure types were associated with a greater bleed incidence on and near the ear.
Among ENT surgery cases, ear surgery has been regarded as low risk of requiring unexpected post operative admission due to complications.219 Applying firm pressure dressings to defects closed on and around the ear often proved challenging.
We have previously identified three additional independent risk factors for bleeding complications following skin surgery by multivariate analysis4; age 67 or older, warfarin administration and closure with flap or graft. The risk of post operative bleeding is 1/1000 with no risk factors, 4/1000 with 1 risk factor and 12/1000 with 2 risk factors. Bleeding risk then escalates with 3 risk factors, (43/1000) and all 4 risk factors, (56/1000).
A blinded study has demonstrated that even experienced surgeons cannot predict whether patients are on warfarin or aspirin at the time of operating. 220
Surgeons should consider anticoagulant management and INR levels prior to operating. This study involves a single surgeon working in two surgical venues in one southern Australian regional city. The bleeding incidence reported may not reflect practice elsewhere.
The only site where post operative bleeding risk was found to have increased risk followed excision on or immediately adjacent to the ear. An increased attention to haemostasis and dressings may be considered when surgery to the ear is planned. Curettage results in a similar bleeding risk to simple surgical excision and closure.
Bleeding following skin procedures is likely to be little more than an inconvenience and manageable with steps such as pressure, dressings and review. Occasionally (1:2000) the patient may come to wound exploration.