Transverse Small Skin Incision for Carotid Endarterectomy

Transverse Small Skin Incision for Carotid Endarterectomy Terezia B. Andr asi,1,2,3 Christof Kindler,2,3 Elke Dorner,2 and Justus Strauch,3 Essen, Mainz, and Bochum, Germany

Background: The purpose of this report is to determine the feasibility of short transverse skin incision (STI < 4 cm) for eversion (EEA) and patch (PEA) endarterectomy with or without shunt by comparing it with the outcomes after long transverse skin incision (LTI 4e8 cm). Methods: Of 164 elective consecutive patients (71 ± 2.73% symptomatic) operated at one institution over 24 months, 81 were treated with STI, while 83 patients received LTI. The LTI and STI groups did not differ in terms of age, symptoms, or risk factors. EEA or PEA under locoregional (LRA) or general (GA) anesthesia were performed. Results: STI was associated with shorter operation times (75.19 ± 15.33 vs. 94.87 ± 41 and 99.4 ± 27.36 vs. 132.66 ± 51.32, respectively, P < 0.01) and similar clamping times (26.05 ± 5.71 vs. 26.07 ± 7.14 and 34.04 ± 9.49 vs. 42.5 ± 20.8, respectively) in the EEA and PEA groups that did not receive shunts compared with the corresponding LTI groups, and the operating room stays of the STI patients operated on GA were shorter than that of the corresponding LTI patients (181.11 ± 39.16 vs. 212.5 ± 64, P < 0.001). Nonsignificant differences were found between the corresponding STI and LTI shunt groups. No perioperative deaths occurred. STI was associated with less perioperative complications than LTI. Macroscopically nondistinguishable scar was present in 85% in the STI and 52% in the LTI groups (P < 0.001). Postoperative local irritation and paresthesia occurred similarly in the STI (11%) and LTI (14%) groups. Conclusions: STIs are feasible for PEA and EEA. STIs produce significantly better cosmetic outcomes and shorter operation times than LTI and have similar rates of complication and similar incidences of local discomfort. Although no neurological consequences of using STIs for PEAs with shunts were revealed, STI should be applied with caution until sufficient patch length and long-term patency of this procedure are demonstrated.

INTRODUCTION Presented at the Annual Meeting of the European Society for Vascular Surgery, Budapest, Hungary, September 2013. 1 Department of Cardiac Surgery, Heart Center Essen-Huttrop, University of Essen, Essen, Germany. 2 Department of Cardiac, Thoracic and Vascular Surgery, Johannes Gutenberg University Clinic, Mainz, Germany. 3 Department of Cardiothoracic Surgery, Bergmannsheil Hospital, Bochum, Germany. Correspondence to: Terezia B. Andrasi, MD, PhD, Department of Cardiac Surgery, Heart Center Essen-Huttrop, Essen University Clinic, Herwarthstr. 100, 45147 Essen, Germany; E-mail: terezia.andrasi@ heh.uk-essen.de

Ann Vasc Surg 2015; 29: 447–456 http://dx.doi.org/10.1016/j.avsg.2014.10.014 Ó 2015 Elsevier Inc. All rights reserved. Manuscript received: June 14, 2014; manuscript accepted: October 19, 2014; published online: November 24, 2014.

The increasing patient demand for short or no hospitalization, rapid recovery, and nondistinguishable and painless postoperative scars has stimulated the implementation of minimally invasive surgical techniques and the use of local anesthesia in the field of carotid repair. Even though minimal invasiveness can easily be achieved via endovascular therapy in the majority of the vascular territories that require treatment, open surgical reconstruction remains the gold standard in reestablishing the function of the carotid artery.1,2 The feasibility of longitudinal small skin incision for carotid endarterectomy under both general 447

448 Andr asi et al.

anesthesia (GA) and local anesthesia has previously been demonstrated.3e5 Patch reconstruction could safely be performed through longitudinal short incision with reduced incidence of nerve injury5 and better cosmetic outcomes.4 In addition to these recent findings, many years ago, Skillman et al.6 reported that, although surgical exposure is simpler when vertical incisions are employed, adequate exposure via transverse incisions is always possible, the incidence of temporary cranial nerve deficits is similar between the 2 incision types, and patients significantly favor transverse incisions. Assadian et al.7 corroborated these findings by demonstrating the feasibility of transverse small skin incision for eversion endarterectomy (EEA) under locoregional anesthesia (LRA). Moreover, the outcomes of small transverse skin incision and vertical incision for carotid endarterectomy under LRA were reported as similar.8 However, the performances of neither carotid endarterectomy with patch reconstruction nor the utilization of indwelling shunts via short transverse skin incisions (STIs) have yet been described. Therefore, the aims of this study were as follows: (1) to determine the feasibility of STI for patch endarterectomy (PEA) and EEA; (2) to determine the influences of anesthesia type and indwelling shunt placement on the operative parameters; (3) to identify factors that may possibly limit the routine use of transverse small skin incision for carotid endarterectomy; (4) and to quantify the incidence of postoperative skin paresthesia and the cosmetic outcomes of small skin incisions compared with those of transverse long skin incisions (LTIs).

MATERIALS AND METHODS Patients From June 2009 to March 2010, 164 consecutive open surgical carotid revascularization procedures were performed at one academic institution. Patients were selected for STI or LTI based on the individual surgeon’s judgment of the presence of a very short, bull-like, or inextensible neck.9 Clinical assessments and neurological and ultrasound (US) evaluations were performed preoperatively. The procedures were performed either under GA or under LRA as described elsewhere.3e5 The selection of the type of anesthesia was at the discretion of the operating surgeon and the anesthesiologist. Intraoperative cerebral oximetric monitoring (INVOS-3100; Somanetics, Troy, Michigan, USA) was performed for patients who were operated on

Annals of Vascular Surgery

under GA. Indwelling shunts placement was determined by a
20% drop in regional cerebral oxygen saturation in the surgeries performed under GA, and based on awake testing and clinical neurological monitoring during carotid clamping in the surgeries performed under LRA. Operative Technique After placement of the patient in Trendelenburg position with the neck in extension, marking of the incision line was performed before sterile draping. A transverse skin incision (4 cm) was made in a neck wrinkle at the level of the upper margin of the cricoid medial to the sternocleidomastoid muscle (Fig. 1A) and offered sufficient exposure of the carotid bifurcation for thromboendarterectomy (Fig. 1B). The STI did not require intraoperative duplex scanning before incision. The use of automatic retractors and the laxity of the skin allowed for traction and exposure of up to 1.5 cm more cranially or caudally located bifurcations. Traction slightly impacted the length of the incision at the end of the operation (Fig. 1C). Schematic representation of the incision lines in correlation with the anatomical landmarks are presented in Figure 1D. Whenever the US scanning described high carotid bifurcation, we placed our transverse skin incision in the next parallel cranial skin wrinkle slightly (a few millimeters) above the level of the upper margin of the thyroid cartilage. Conversion into long transversal incision could be easily performed through latero-postero-cranial prolongation of the transverse incision line. After surgical exposure, the arterial branches were surrounded with vessel loops prior to carotid cross-clamping (Fig. 1B). A thin Langenbeck retractor was used to perform elevation rather than cranial traction of the soft tissues that enabled visualization of the internal carotid artery over a length of up to 2 cm under the skin. Small bulldog clamps could be placed deeply into the wound beyond the extent of the skin incision for distal clamping. These maneuvers allow the length of the exposed carotid arteries to be longer than the length of the skin incision. Endarterectomies of the internal carotid, external carotid, common carotid and, when necessary, superior thyroid arteries were performed through either longitudinal incision or transection based on intraoperative findings and the surgeon’s discretion. Longitudinal incisions were closed with bovine pericardium patches (Vascutek Terumo, Inchinnan, Scottland, UK), and the eversion technique was followed by direct end-to-side reimplantation.

Vol. 29, No. 3, April 2015

Transverse small skin incision for carotid surgery

449

Fig. 1. Short transverse skin incision for carotid endarterectomy. (A) Marking the incision before the operation. (B) Surgical exposure of the carotid arteries and the superior thyroid artery before carotid cross-

clamping. (C) Wound length after skin closure. (D) Schematic representation of the transverse small skin incision line in comparison with the transverse long and longitudinal incisions.

Postoperative Care

quantified. Conversions of small transverse skin incisions (<4 cm) into long transverse incisions (>5 cm) were documented. Postoperative complications, such as bleeding and nerve damage, were evaluated.

All patients who were operated on under GA were extubated in the operation room immediately following the operation. Hemodynamic monitoring was employed for the first 24 postoperative hours. Blood pressure was adjusted when necessary. Drainage was removed on the first postoperative day. Full motion was allowed 12 hr after the operation. Patients were discharged when ambulation was independent, cognitive function was intact, and blood pressure and glucose levels were controlled. Data Evaluation The demographic data, neurological features, perioperative details, complications, cosmetic results (macroscopically nondistinguishable postoperative scars) at 6 months self-assessed by each patient and presence of paresthesia up to 6 month were retrospectively reviewed. The operation room times (defined as the time from arrival of the patient into the operating room until the patient’s exit from the operating room), operating times (defined as the time from skin incision to skin closure), and clamping/shunting times were

Statistical Analyses The data are presented as means ± standard deviations and as counts (percentage) where appropriate. The demographic homogeneities of groups containing at least 7 patients were tested. The type of incision (STI versus LTI) was compared between groups (n > 6) in terms of anesthesia type (GA versus LRA), surgical technique (EEA versus PEA), and carotid shunting (with shunt versus without shunt). Statistical evaluations of the clinical and demographic characteristics were performed with chi-squared tests. ManneWhitney U tests were used to compare between-group data. Wilcoxon tests were used to compare the linear changes in each group. The SPSS 5.0 (SPSS Inc., Chicago, IL, USA) and Origin Pro (Origin Pro, Northampton, MA, USA) programs were used for analyses. Statistical significance was interfered at P < 0.05.

450 Andr asi et al.

Annals of Vascular Surgery

Table I. Flow chart of the patient collective Incision

Total number of patients 164 LTI 83

Anesthesia Endarterectomy Without shunt With shunt

STI 81

GA 73 (88%) PEA 35 9 26y

EEA 38 37 1

LRA 10 (12%) PEA 1 1 0

EEA 9 7 2

GA 58 (72%) PEA 32 16 16

EEA 26 22 4

LRA 23 (28%)* PEA 14* 7 7

EEA 9 8 1

*P < 0.01 compared with the corresponding LTI subgroup. y P < 0.01 compared with the corresponding PEA subgroup without shunt.

RESULTS Patient Population LTI and STI techniques were employed for 83 and 81 of the total 164 patients, respectively (Table I). A significantly greater number of patients (P < 0.05) in the STI group (28%) were operated on under LRA than in the LTI group (12%). The numbers of patients who received similar surgical procedures were comparable in all respective subgroups (STI versus LTI) with the exception of reconstruction under LRA, which was performed more often in the small incision than the long incision groups (14 vs. 1, respectively, P < 0.05). Of the patients who required temporary shunts, 74% received PEA under GA, and these patients were more likely to have received a long incision (45%) than a short incision (28%, P < 0.05). All the other subgroups were comparable in terms of the numbers of patients. Demographic Data The demographic data of all patients are presented in Table II. Only the groups marked in gray (n > 6) were statistically compared. The patients who received short skin incisions were predominantly men, and the ages of the LTI and STI groups were similar. The patients who received EEAs through STIs under LRA were less adipose (P ¼ 0.02) and less likely to suffer from diabetes mellitus (P ¼ 0.02) than the patients who received the same procedure through LTIs. In contrast, the patients who received STIs and required an intraoperative shunt for PEA were more adipose than those who underwent a similar treatment through an LTI (P ¼ 0.001). There were no differences

between the corresponding groups that were operated on under GA without intraoperative shunts. Preoperative Neurological Status A comparison of the total numbers of patients operated on with and without shunts (Table III, first column) revealed that only 42 (37.17%) of the 113 symptomatic patients required the placement of an indwelling shunt (P < 0.001). The incidence of stroke did not differ (P ¼ 0.261) between the groups that received operations with and without shunt; however, a significantly higher percentage of patients with contralateral internal carotid artery (ICA) stenosis were found in both the LTI and STI groups (P < 0.01) compared with the nonshunt groups. Homogenous distributions of patients were observed in the 10 corresponding comparable subgroups (n > 6) that received EEA or PEA with or without shunts under GA or LRA (Table III, first and second blocks of columns). The mean degrees of stenosis on the operated side varied between 72% and 95% (range: 70% and 98%). The use of an intraoperative temporary shunt was not influenced by the type of anesthesia (30.33% in the LRA vs. 35.87% in the GA groups, P ¼ 0.548) or by the type of skin incision (34.56% in the STI vs. 34.93% in the LTI groups, P ¼ 0.96; Table III, last block of columns) and was associated with similar incidence of perioperative stroke than the nonshunt groups. Perioperative Times The operating room times and operating times (Fig. 2) were significantly shorter (P < 0.001 and P < 0.0001, respectively) among all patients who

Vol. 29, No. 3, April 2015

Transverse small skin incision for carotid surgery

451

Table II. Patient characteristics GA

Incision

LRA

PEA

PEA

EEA

EEA

PEA

PEA

EEA

EEA

LTI

STI

LTI

STI

LTI

STI

LTI

STI

Without shunt (n) 9 16 37 22 1 7 7 8 Age (years) 70.00 ± 5.5 70.62 ± 8.1 71.56 ± 9.1 70.79 ± 10.5 67.00 74.14 ± 5.6 67.43 ± 9.7 66.67 ± 10.6 Male (%) 77.70 87.50 70.27 54.54 100 85.71 100.00 100.00 HLP (%) 33.33 31.25 56.75 45.45 0 28.57 28.57 62.50 DM (%) 66.66 31.25 27.02 40.90 100 42.85 71.42 12.50* AHT (%) 100.00 87.50 100.00 68.18 100 85.71 57.14 87.50 Adiposity (%) 33.33 25.00 48.64 27.27 0 57.14 71.42 12.50* Nicotine (%) 66.66 56.25 24.32 27.27 100 42.85 28.57 12.50 With shunt (n) 26 16 1 4 0 7 2 1 Age (years) 71.27 ± 6.7 72.37 ± 8.9 69 73.75 ± 8.2 e 77.14 ± 6.4 72.00 ± 7.0 79 Male (%) 73.07 75.00 100 50.00 e 85.71 100 100 HLP (%) 53.85 56.25 0 25.25 e 57.14 100 100 DM (%) 23.07 43.75 0 0 e 42.85 100 0 AHT (%) 46.14 73.33 0 100.00 e 100.00 100 100 Adiposity (%) 19.23 68.75* 0 0 e 42.85 0 0 Nicotine (%) 23.07 43.75 0 0 e 42.85 100 100 Age is expressed as mean + SD, incidences of male gender, adiposity and nicotine consume are given in percentage (%), and n represents the number of patients in each group. AHT, arterial hypertension; DM, diabetes mellitus; HLP, hyperlipoproteinemia. *P < 0.01 for STI versus corresponding LTI group.

underwent STI (n ¼ 81) compared with all patients who underwent LTI (n ¼ 83). LRA was associated with significantly shorter operating room times than GA (Fig. 3). STIs were associated with shorter operating room times than LTIs under GA (P < 0.001); however, no differences between STI and LTI occurred under LRA (P ¼ 0.726). In terms of operative technique (Fig. 3), EEA operation times were significantly shorter than those of PEA regardless of whether long skin incisions (P ¼ 0.001) or short skin incisions (P < 0.001) were used. Clamping times were not influenced by the type of skin incision or by the type of anesthesia (Fig. 4). Comparison of all subgroups (Fig. 5) revealed that under GA EEA through STI required a shorter operation time than did EEA through LTI (P ¼ 0.024); however, this effect was not assessed in the PEA group where operation times were similar in the STI and LTI subgroups (P ¼ 0.08). Surprisingly, under LRA, the mean operation times of both the EEA and PEA groups were longer following STI than LTI, but these differences did not reach significance (P ¼ 0.22, STI versus LTI and EEA groups with n > 6). However, the EEA operation times were significantly shorter than the PEA operation times, in the corresponding groups without shunts when short skin incisions were applied under both

GA (P ¼ 0.02) and LRA (P ¼ 0.007). Indwelling shunt placement did not influence the operation times of any of the corresponding groups. PEAs with shunts performed under GA (Fig. 6) required significantly shorter clamping times and shorter shunt times through STIs than through LTIs (P ¼ 0.003 and P ¼ 0.002, respectively), while EEAs performed without shunts under LRA required significantly longer clamping times through STIs than through LTIs (P ¼ 0.004). Complications Three conversions of STIs into LTIs were necessary because of undiagnosed poststenotic aneurysm of the ICA with parietal thrombosis, high carotid bifurcation, and long calcification of the ICA. No postoperative complications occurred because of these conversions. One ipsilateral stroke and 2 nerve injuries (1 recurrent laryngeal nerve and 1 superior laryngeal nerve) occurred in the PEAeGAeLTI group, 2 incidences of bleeding occurred in the EEAeGAeLTI group, and 1 incidence of bleeding occurred in the EEAeGAeSTI group. The complication rates of the STI and LTI groups were not different (P ¼ 0.102). No perioperative deaths occurred.

452 Andr asi et al.

Table III. Preoperative neurological features GA

Incision

Without shunt (n) Degree of stenosis Symptomatic patients Stroke Patients with contralateral stenosis With shunt (n) Degree of stenosis Symptomatic patients Stroke Patients with contralateral stenosis

LRA

Total

PEA

PEA

EEA

EEA

PEA

PEA

EEA

EEA

LTI

STI

LTI

STI

LTI

STI

LTI

STI

GA

LRA

P value LTI

STI

P value

16

37

22

1

7

7

8

84

23

0.001 54

53

0.891

85%

91%

95%

95%

80%

88%

70%

91%

80%

e

86%

e

107 9 95%

91%

71 5 (55%) 9 (56%) 29 (78%) 13 (59%) 0 (0%) 6 (88%) 4 (57%) 5 (63%) 56 (66%) 15 (65%) 0.891 38 (70%) 33 (62%) 0.375 (e) (e) 9 0 3 2 3 0 1 0 0 8 1 0.426 2 7 0.044 6 0 0 3 1 0 1 1 0 4 2 0.468 4 2 0.500

57 26 88%

16 84%

1 90%

4 85%

0 e

42* 22 (85%) 11 (69%) 1 (100%) 2 (50%) e 8 4 17* 8

1 5y

0 0

0 3

e

7 89%

2 80%

1 72%

47 86%

10 86%

0.001 29 e 87%

28 85%

0.851 e

4 (57%) 1 (50%) 1 (100%) 36 (76%) 10 (100%) 0.089 24 (82%) 18 (64%) 0.044 (e) (e) 2 1 0 5 3 0.234 5 3 0.734 1 0 0 16* 1 0.131 8* 9* 0.707

Annals of Vascular Surgery

Degree of stenosis is given in % after NASCET; numbers represent number of patients; % in brackets represents percentage of symptomatic patients of each subgroup. P, value of P when comparing parameters of LRA versus GA groups and parameters of STI versus LTI groups. *P < 0.01 compared with the corresponding group without shunt. y P < 0.05 compared with the corresponding group without shunt.

Vol. 29, No. 3, April 2015

Fig. 2. Operation room times and operating times in groups with short (STI) to long (LTI) skin incision.

Cosmetic Results and Local Paresthesia Macroscopically nondistinguishable postoperative scars were documented in 69 of the patients in the STI group (85%) and 43 of the patients in the LTI group (52%) 6 months after the operations; this difference was statistically significant (P < 0.001). The incidences of postoperative local irritation and numbness or paresthesia during shaving lasting up to 6 months were similar in the STI (11%) and LTI (14%) groups (P ¼ 0.717).

DISCUSSION This study revealed that STI offers significant cosmetic benefits, does not jeopardize the quality of the carotid endarterectomy in terms of perioperative time and complication rate regardless of the surgical technique, and is feasible under both GA and LRA. Homogenous distributions of risk factors were found in all subgroups, and, surprisingly, adiposity was not a contraindication for the use of STI in patients who required PEAs with shunts (Table II). Together these data indicate that demographic factors are of minimal importance when deciding on the length of skin incision for carotid endarterectomy. In symptomatic patients requiring a shunt, STI was employed less frequently. We found that the need for a shunt was determined more by the presence of significant contralateral disease than by the patient’s presenting symptoms (Table III). This finding is consistent with previous data that have shown that collateral brain circulation is the predominant determinant of the need for intraoperative shunting.10 However, our data additionally demonstrated that the above-mentioned neurological factors do

Transverse small skin incision for carotid surgery

453

not limit the use of short skin incision or local anesthesia when indwelling shunt placement is not necessary. In agreement with the results of the GALA trial,11 our data indicated no drawbacks of local anesthesia in terms of postoperative complications and surgical and cosmetic outcomes. LRA could be successfully applied regardless of demographic factors, preoperative neurological statuses, or indwelling shut placements in both the STI and LTI groups (Tables II and III). Moreover, we revealed that LRA required significantly reduced operating room time compared with GA after both LTIs and STIs (Fig. 3). Regarding the types of surgical techniques, our study showed that EEA was associated with significantly shorter operating times than was PEA following both long and short skin incisions (Fig. 3). Moreover, STI additionally reduced operating time compared with LTI for both EEA and PEA without increasing the complication rates. These findings do not contradict previous studies that reported that better early outcomes are associated with EEA12 or that the late outcomes of EEA and PEA are similar.13 Contrarily, Demirel et al.14 found that intraoperative and early postoperative neurologic complications occur more frequently after EEA. In the absence of early complications, however, the EEA and PEA may yield similar long-term patency. One explanation could be that although EEA is not accompanied by an increase in arterial diameter and the operated carotid artery resembles an unoperated, nonstenotic artery, the resulting turbulent flow is not significantly different than that associated with PEA.15,16 Nevertheless, the only case of stroke noticed in our study occurred in the PEA group. In contrast, other investigators17 found that PEA was associated with a lower rate of perioperative stroke and was superior in lowering the incidence of restenosis. However, reduced restenosis rate did not appear to be associated with clinical benefit in terms of reduced stroke risk.18 The rate of the total complication rate in our study was similar in the STI and LTI groups. The postoperative bleeding that occurred in one patient in the STI group might have been related to difficult examinations of the dorsal anastomotic segments during EEA after reperfusion. We also believe that the nerve damage observed in 2 patients in the LTI group were presumably caused during the more extensive surgical exposure required for PEA.8,19 In the cases with extensive plaques requiring patch reconstruction, although distal clamping and endarterectomy could be easily achieved, difficulties to place the stitches at the cranial end of the patch reconstruction were sometimes assessed, even

454 Andr asi et al.

Annals of Vascular Surgery

Fig. 3. Influence on the type of anesthesia on the operation room time and influence on the type of surgical technique on the operating time. *P < 0.05, LRA versus GA and EEA versus PEA corresponding groups.

Fig. 4. Clamping time and shunting time of the internal carotid artery in groups with short (STI) to long (LTI) skin incision. Clamping times, clamping times before and after

shunting, and shunting times of the internal carotid artery are presented. P < 0.05 is significant.

though parachute technique was always used. Necessity of additional stitches at the cranial end of the anastomosis and improper placement could further increase the risk of late stenosis at the distal end of the anastomosis. However, only one case in our study required conversion to long incision

because of an extensive plaque and no postoperative complications occurred. We consider that preoperative preselection based on plaque length is not necessary, because intraoperative extension of the incision can be performed without increasing the risk for complications.

Vol. 29, No. 3, April 2015

Transverse small skin incision for carotid surgery

455

Fig. 5. Operation time in each subgroup.

Fig. 6. Clamping times and clamping with shunting times in each subgroup.

The significantly shorter operation times observed for the PEA groups that received STI compared with those that received LTI should be interpreted with caution because we did not measure the length of the patches during PEA. One possible explanation for the shorter operative times in some patients undergoing STI is that the short incisions might force the surgeon to make a shorter arteriotomy and thus implant a shorter patch, which obviously requires less time than implantation of a longer patch; however, no residual stenosis was found in the early postoperative screenings, and no revisions for residual stenosis were necessary. In contrast, the

clamping times were similar between the STI and LTI groups under both GA and LRA (Fig. 4), which suggests that similar times were required for arterial desobliteration and surgical reconstruction. Although the comparison of the subgroups (Fig. 5) revealed no difference between the LTI and STI groups when PEAs without shunts were performed, significantly lower shunt times were found for STI than for LTI when temporary shunting during PEA was required (Fig. 6). This latter finding cannot exclude the possibility that patch length is shorter with STI. The use of indwelling shunts during EEA was associated with similar operation times in the

456 Andr asi et al.

STI and LTI groups. Moreover, clamping times were also similar in the STI and LTI groups under both GA and LRA, which suggest that the times spent on desobliteration and performing the anastomosis were comparable across all EEA groups. Another limitation of our study is that it did not quantify in detail injuries of the cervical nerves. In a recent study on 228 patients, Mendes et al.19 revealed 3% incidence of cranial nerve injury in the transverse incision group and 5.3% in the longitudinal incision group. All injuries were transient and the authors concluded that transverse skin incision is not associated with an increased risk of cerebral nerve injury. Another recently published meta-analyses of 6,878 patients undergoing carotid surgery20 revealed a 5.6% total incidence of cranial nerve injury. Predictors for cerebral nerve injury were urgent procedures, immediate reexploration after closure under the same anesthetic, and return to the operating room for a neurologic event or bleeding. The low incidence of surgical reexploration in our study and exclusion of emergency cases could justify the low incidence of nerve injury rate in the presented patient population. Moreover, we experienced shorter operation times associated to the use of short skin incision. Accordingly, in a detailed study, Bastounis et al.5 concluded that longer operating times are associated with a significantly greater incidence of cervical nerve damage. Last but not the least, although in our study STI required significantly shorter operation times than did LTI (Fig. 2), the rates of postoperative skin paresthesia were similar in these groups. In conclusion, STIs can be applied for both PEA and EEA regardless of risk factors. STIs offer significantly better cosmetic outcomes and shorter operation times than LTIs, and have similar rates of complications and similar incidences of local discomfort as LTIs. In terms of overall outcome, EEA under LRA was found to be the most feasible technique among the procedures performed through STI. Although no neurological consequences of the use of STIs for PEAs with shunts were revealed, long incisions might be safer until imaging studies confirm flawless patch length, and the long-term patency of this surgical procedure is assured.

Annals of Vascular Surgery

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

REFERENCES 1. Sidawy AN, Zwolak RM, White RA, et al. Outcomes Committee for the Society for Vascular Surgery. Risk-adjusted 30-day outcomes of carotid stenting and endarterectomy: results from the SVS Vascular Registry. J Vasc Surg 2009;49:71e9. 2. Ricotta JJ 2nd, Malgor RD. A review of the trials comparing carotid endarterectomy and carotid angioplasty

19.

20.

and stenting. Perspect Vasc Surg Endovasc Ther 2008;20: 299e308. Ascher E, Hingorani A, Marks N, et al. Mini skin incision for carotid endarterectomy (CEA): a new and safe alternative to the standard approach. J Vasc Surg 2005;42:1089e93. De Troia A, Miosso F, Biasi L, et al. Carotid endarterectomy with mini-invasive access in locoregional anaesthesia. Acta Biomed 2008;79:123e7. Bastounis E, Bakoyiannis C, Cagiannos C, et al. A short incision for carotid endarterectomy results in decreased morbidity. Eur J Vasc Endovasc Surg 2007;33:652e6. Skillman JJ, Kent KC, Anninos E. Do neck incisions influence nerve deficits after carotid endarterectomy? Arch Surg 1994;129:748e52. Assadian A, Senekowitsch C, Pfaffelmeyer N, et al. Incidence of cranial nerve injuries after carotid eversion endarterectomy with a transverse skin incision under regional anaesthesia. Eur J Vasc Endovasc Surg 2004;28:421e4. Marcucci G, Antonelli R, Gabrielli R, et al. Short longitudinal versus transverse skin incision for carotid endarterectomy: impact on cranial and cervical nerve injuries and esthetic outcome. J Cardiovasc Surg 2011;52:145e52. Frego M, Bridda A, Ruffolo C, et al. The hostile neck does not increase the risk of carotid endarterectomy. J Vasc Surg 2009;50:40e7. Kim GE, Cho YP, Lim SM. The anatomy of the circle of Willis as a predictive factor for intra-operative cerebral ischemia (shunt need) during carotid endarterectomy. Neurol Res 2002;24:237e40. GALA Trial Collaborative Group. Lewis SC, Warlow CP, Bodenham AR, et al. General anaesthesia versus local anaesthesia for carotid surgery (GALA): a multicentre, randomised controlled trial. Lancet 2008;372:2132e42. Antonopoulos CN, Kakisis JD, Sergentanis TN, et al. Eversion versus conventional carotid endarterectomy: a meta-analysis of randomised and non-randomised studies. Eur J Vasc Endovasc Surg 2011;42:751e65. Cao P, Giordano G, De Rango P, et al. Eversion versus conventional carotid endarterectomy: late results of a prospective multicenter randomized trial. J Vasc Surg 1998;27:595e605. Demirel S, Macek L, Bruijnen H, et al. Eversion carotid endarterectomy is associated with decreased baroreceptor sensitivity compared to the conventional technique. Eur J Vasc Endovasc Surg 2012;44:1e8. Baan J Jr, Thompson JM, Reul GJ, et al. Vessel wall and flow characteristics after carotid endarterectomy: eversion endarterectomy compared with Dacron patch plasty. Eur J Vasc Endovasc Surg 1997;13:583e91. Crawford RS, Chung TK, Hodgman T, et al. Restenosis after eversion vs patch closure carotid endarterectomy. J Vasc Surg 2007;46:41e8. AbuRahma AF, Robinson PA, Saiedy S, et al. Prospective randomized trial of carotid endarterectomy with primary closure and patch angioplasty with saphenous vein, jugular vein, and polytetrafluoroethylene: long-term follow-up. J Vasc Surg 1998;27:222e32. Cao PG, de Rango P, Zannetti S, et al. Eversion versus conventional carotid endarterectomy for preventing stroke. Cochrane Database Syst Rev 2001;(1):CD001921. Review. Mendes GA, Zabramski JM, Elhadi AM, et al. Carotid endarterectomy: comparison of complications between transverse and longitudinal incision. Neurosurgery 2014;75:110e6. Fokkema M, de Borst GJ, Nolan BW, et al. Clinical relevance of cranial nerve injury following carotid endarterectomy. Eur J Vasc Endovasc Surg 2014;47:2e7.