Transverse versus vertical groin incision for femoral artery approach

Abstract

Background

Access to the femoral vessels is necessary for a wide range of vascular procedures, including treatment of thromboembolic disease, arterial grafts (i.e. bifemoral aortic bypass or infrainguinal bypass), endovascular repair of abdominal aortic aneurysm (EVAR), thoracic endovascular aneurysm repair (TEVAR) and transcatheter aortic valve implantation (TAVI). The surgical technique used to access the femoral artery may be a factor in the occurrence of postoperative complications; this will be the focus of our review. We will compare the transverse surgical technique—a cut made parallel to the groin crease—versus the vertical groin incision surgical technique—classic technique: a surgical cut made across the groin crease—to access the femoral artery, in an attempt to determine which technique has the lower rate of complications, is safer and is more effective.

Objectives

To evaluate the efficacy and safety of transverse groin incision compared with vertical groin incision for accessing the femoral artery in endovascular surgical procedures and open surgery.

Search methods

The Cochrane Vascular Information Specialist searched the Cochrane Vascular Specialised Register, CENTRAL, MEDLINE, Embase, CINAHL and AMED databases, and the World Health Organization (WHO) International Clinical Trials Registry Platform and ClinicalTrials.gov to 17 February 2020. The review authors searched the IBECS database to 26 March 2020 and reference lists of relevant studies/papers.

Selection criteria

We included randomized controlled trials (RCTs) and quasi‐randomized trials (qRCTs) that compare transverse and vertical groin incision, during either endovascular or open surgery procedures.

Data collection and analysis

Two review authors (MVCRC, FCN) independently selected the studies, assessed risk of bias, extracted data, performed data analysis and graded the certainty of evidence according to GRADE.

Main results

We included one RCT and one qRCT in this review. These two studies had a combined total of 237 participants (283 groins). Infection of the surgical wound was the only outcome that was similar in both studies, and that could therefore be submitted to a combined analysis.

Meta‐analysis of the two studies showed low‐certainty evidence that transverse groin incision resulted in a lower risk of surgical wound infection in the 10‐ to 28‐day period following surgery (risk ratio [RR] 0.25, 95% confidence interval [CI] 0.08 to 0.76; 2 studies; 283 groin incisions). There was low heterogeneity between the studies. We downgraded the certainty of the evidence for surgical wound infection by one level due to serious limitations in the design (there was a high risk of bias in critical domains). The confidence interval for surgical wound infection is relatively wide, further indicating that the certainty of the effect estimate is low. This is likely due to the small number of studies and participants. We observed no evidence of a difference between the two surgical techniques for the other evaluated primary outcome 'lymphatic complications': lymphocele (RR 0.46, 95% CI 0.20 to 1.02; 1 study; 116 groins); and lymphorrhea (RR 2.77, 95% CI 0.92 to 8.34; 1 study; 116 groins). We downgraded the certainty of evidence for lymphatic complications by one level due to serious limitations in the design (there was a high risk of bias in critical domains); and by two further levels because of imprecision (small number of participants and only one study included).

High‐quality studies are needed to enable a comparison of the two surgical techniques with respect to other outcomes, such as infection of the vascular graft (endoprosthesis/prosthesis), prolonged hospitalization, reoperative surgery, death, neurological deficit (e.g. paresthesia), amputation, graft patency, and postoperative pain.

Authors' conclusions

In this systematic review, we found low‐certainty evidence that performing transverse groin incision to access the femoral artery resulted in fewer surgical wound infections compared with performing vertical groin incision. We observed no evidence of a difference between the two surgical techniques for the other evaluated outcomes (lymphocele and lymphorrhea). Other outcomes were not evaluated in these studies.

Limitations of this systematic review are, however, the small sample size, short clinical follow‐up period and high risk of bias in critical domains. For this reason, the applicability of the results is limited.

Plain language summary

Transverse versus vertical incisions at the inguinal (groin) region for femoral artery approach

Background

Access to the femoral vessels is necessary for a wide range of vascular procedures, including treatment of thromboembolic disease, arterial grafts, endovascular repair of abdominal aortic aneurysm, thoracic endovascular aneurysm repair and transcatheter aortic valve implantation. The surgical technique used to access the femoral artery may be a factor in the occurrence of postoperative complications; this is the focus of our review.

We compared the transverse surgical technique (a cut made parallel to the groin crease) versus the vertical groin incision at the inguinal region (classic surgical technique: a cut made across the groin crease) to access the femoral artery, in an attempt to determine which technique has the lower rate of complications, is safer and is more effective.

Study characteristics and key results

This systematic review includes two studies (most recent search February 2020): one randomized controlled trial and one quasi‐randomized clinical trial. Both compared transverse versus vertical inguinal approaches. One study included 149 participants (167 groins) while the second study included 88 participants (116 groins), undergoing inguinal surgery to access the femoral artery.

The outcome 'wound or surgical site infection' was assessed in both studies. The combined analysis showed a lower rate of wound infections for the transverse inguinal incision compared with the vertical inguinal incision. One study assessed lymphatic complications and found no evidence of a difference between the two incision techniques. Other outcomes such as infection of the graft, hospitalization, death and postoperative pain were not reported by the two studies

Certainty of the evidence

We classified the certainty of the evidence as low for surgical site infection due to the high risk of bias because of issues with randomization and the blinding of people assessing the outcomes and the small number of participants in included studies. We considered the lymphatic complications of very low certainty evidence due to the high risk of bias because of issues with randomization and the blinding of people assessing the outcomes, and because there was only one included study with a small number of participants assessing lymphatic complications.

Conclusion

Evidence of low certainty suggests that surgical wound infection in the 28‐day period post surgery occurs less frequently in transverse incisions than in vertical incisions to access the femoral artery. Evidence of very‐low certainty indicated that there was no evidence of a difference between the two surgical techniques relating to the lymphatic complications' outcome for access to the femoral artery in the 28‐day period post surgery.

Summary of findings

Summary of findings for the main comparison. Transverse groin incision versus vertical groin incision for femoral artery approach during vascular procedures.

Vertical groin incision versus transverse groin incision for femoral artery approach during vascular procedures
Patient or population: patients undergoing surgery with femoral artery approach Settings: hospital Intervention: transverse groin incision Comparison: vertical groin incision
Outcomes	Anticipated absolute effects* (95% CI)		Relative effect (95% CI)	No. of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with control (vertical groin incision)	Risk with intervention (transverse groin incision)
Surgical site infection (10 to 28 days)	Study population		RR 0.25 (0.08 to 0.76)	283 (1 RCT; 1 qRCT)	⊕⊕⊝⊝a, b low
	103 per 1000	26 per 1000 (8 to 78)
Lymphatic complications: Lymphatic leak (28 days) Lymphatic collection (28 days)	Study population		RR 0.46 (0.20 to 1.02)	116 (1 qRCT)	⊕⊝⊝⊝,c,d very low
	279 per 1000	128 per 1000 (56 to 284)
	Study population		RR 2.77 (0.92 to 8.34)	116 (1 qRCT)	⊕⊝⊝⊝c,d very low
	66 per 1000	182 per 1000 (60 to 547)
Infection of the vascular graft (endoprosthesis /prosthesis) (early, ≤ 4 months; late, > 4 months)	‐		‐	‐	‐	The studies included in this review did not report on this outcome
	‐	‐
Prolonged hospitalization (> 1 week)	‐		‐	‐	‐	The studies included in this review did not report on this outcome
	‐	‐
Reoperative surgery (≤ 30 days; > 30 days)	‐		‐	‐	‐	The studies included in this review did not report on this outcome
	‐	‐
Death	‐		‐	‐	‐	The studies included in this review did not report on this outcome
	‐	‐
Neurological deficit (e.g. paresthesia)	‐		‐	‐	‐	The studies included in this review did not report on this outcome
	‐	‐
*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; qRCT: quasi‐randomized controlled trial; RCT: randomized controlled trial; RR: risk ratio
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect

^aHigh risk of bias in random sequence generation (selection bias), blinding of participants and personnel (performance bias), blinding of outcome assessment (detection bias), selective reporting, and incomplete outcome data (attrition bias); downgraded by 1 level (methodological limitation).
 ^b Small number of participants and studies (doubt about the reproducibility of the data); downgraded by 1 level (imprecision).
 ^cHigh risk of bias in random sequence generation (selection bias), blinding of participants and personnel (performance bias), blinding of outcome assessment (detection bias); downgraded by 1 level (methodological limitation).
 ^dSmall number of participants and 1 single study (doubt about the reproducibility of the data); downgraded by 2 levels (imprecision).

Background

Description of the condition

Access to the femoral vessels is necessary for a wide range of vascular procedures, including treatment of thromboembolic disease, arterial grafts (i.e. bifemoral aortic bypass or infrainguinal bypass), endovascular repair of abdominal aortic aneurysm (EVAR), thoracic endovascular aneurysm repair (TEVAR) and transcatheter aortic valve implantation (TAVI). A growing number of vascular procedures using access to the femoral artery are conducted annually; however, the rate of complications related to femoral artery access is still high. The surgical technique used to access the femoral artery may be a factor in the occurrence of postoperative complications; this is the focus of our review. We compare femoral artery access via the transverse surgical technique versus the vertical groin incision surgical technique (classic technique [Swinnen 2010] [Figure 1]), in an attempt to determine which technique has the lower rate of complications, is safer and is more effective.

1.

Vertical and transverse groin incision (image created by M Canteras)

With the development of endovascular procedures observed in recent years, such as those conducted to repair abdominal aortic aneurysms (AAA), the number of surgical procedures to access the femoral vessels has grown. This surgery is performed in the inguinal (groin) region, located in the upper third of the thigh and known as the femoral triangle. Between 2005 and 2008, 84% of aneurysm corrections were performed by EVAR, compared with 42.2% between 1996 and 2002 (Albuquerque 2010). A recent study found similar statistics, namely a 79% rate of aneurysm corrections performed by EVAR, including both ruptured and non‐ruptured cases (Schmitz‐Rixen 2017).

In many of these surgical procedures, percutaneous sealing devices are used. These devices are capable of percutaneously occluding the opening made in the artery without requiring open surgery to suture it (which involves the surgical repair of the entry point in the femoral artery and subsequent closure of the cut‐down incision) (Gimzewska 2017). These devices enable hemostasis even in complex procedures such as EVAR, TEVAR or TAVI, for which high‐end, high‐caliber devices are required. An equivalent increase has been observed in the rate of complications related to surgical access, e.g. surgical site infection, leading to a marked increase in the costs incurred by the healthcare system. Vascular surgical site infection is estimated to add USD 10,000 to the cost of care per patient (Murphy 2015).

Surgical complications associated with femoral artery access include soft tissue infection, lymphatic complications (lymphatic leakage, lymphocele), and prosthesis infection. Other less frequent complications include compromised lower graft patency, neurological injury (paresthesia), need for re‐operation, limb amputation and pain (Beirne 2008). These complications are often difficult to treat and have resulted in prolonged hospital stay (2 to 120 days with lymphatic complications), late rehabilitation, need for additional surgery, increased hospital expenses, higher morbidity and death (Caiati 2010; Shermak 2005).

As a strategy to reduce the risk of surgical infection, the Centers for Disease Control and Prevention created the Surgical Wound Classification (CDC 2018), which divides surgical wounds into four classes—clean, clean‐contaminated, contaminated, and infected—whereby the more advanced the class, the greater the risk of surgical infection. By this definition, lower limbs with lesions/wounds are classified as contaminated or infected, and surgery conducted in patients with such lesions is more hazardous as the pre‐existence of wounds increases the risk of surgical infection.

There is, however, no consensus in the literature about which surgical technique is associated with a lower rate of complications. The aim of the present systematic review is therefore to compare the effectiveness and safety of the transverse groin incision surgical technique with vertical groin incision for accessing the femoral artery.

Description of the intervention

Both the transverse and vertical surgical incision techniques require that the patient lies in a supine position with a slight lateral rotation of the lower limb (i.e. on their back with the leg turned outwards). This position relaxes the musculature of the inguinofemoral region and maximizes access to the femoral triangle.

The femoral triangle (or trigone) is bounded by the inguinal ligament superiorly, the long adductor muscle medially, and the medial border of the sartorius muscle laterally. Its floor is formed by the iliopsoas and pectineus muscles, and its roof is formed by the fascia lata. The femoral triangle contains the femoral nerve, femoral artery and femoral vein (arranged in this order, in the lateromedial direction) (Rossi 2014).

The surgical procedure begins with a skin incision, as limited as possible in length to minimize operative wound size and related complications. Access to the femoral artery can then be achieved in either of two ways.

• The classic vertical technique to access the femoral artery, involving an incision joining the midpoint of a line extending from the superior anterior iliac spine to the tuberosity of the pubis and the midpoint of a line extending from the superior anterior iliac spine to the medial condyle of the femur.

• The transverse groin incision technique to access the femoral artery, involving an incision performed parallel to and below the inguinal ligament, which is used as an anatomical landmark.

After skin incision, superficial lymph vessels and lymph nodes should be ligated and sectioned to avoid lymphatic complications, such as lymphorrhea and lymphocele. Then the superficial lymphatic vessels should be displaced medially towards the pubis to access deeper planes and facilitate access to the femoral vessels. The deep (or enigmatic) fascia that lines the femoral vessels should then be incised vertically (most surgeons perform this surgical technique) or in a slightly arched manner (in both surgical techniques), to produce a medial concavity that exposes the femoral vessels contained in the femoral sheath (Rossi 2014).

Finally, the femoral artery is dissected to the extent that allows its clamping, and subsequent performance of the surgical procedure of choice, such as arterial grafts, bifemoral aortic bypass or infrainguinal bypass, thromboembolectomy, EVAR, TEVAR, or TAVI.

The aim of this systematic review is to assess whether the transverse groin incision surgical technique on patients undergoing surgery involving access to the femoral artery is more effective and safer than vertical groin incision.

How the intervention might work

Although both surgical techniques are performed routinely, surgeons' opinions are divided as to which surgical technique has a lower rate of complications. Transverse groin incision has the following potential advantages over vertical incisions.

• The skin incision is performed following Langer lines (lines of skin tension that correspond to the direction of the elastic fibers and of the skin folds). This promotes reduced tension on the wound edges, thus contributing to a more favorable healing process and reducing the risk of suture dehiscence (Swinnen 2010).

• The skin and subcutaneous tissue are divided according to a more favorable pathway, causing less vascular injury and providing greater oxygen tension to the skin (compared with vertical groin incision), thereby contributing to improved healing (Caiati 2010).

• Access to the femoral vessels is achieved without crossing the groin crease (a region susceptible to greater bacterial colonization), thus reducing the risk of surgical wound infection (Caiati 2010).

• A good surgical field is provided for access to the femoral vessels (Caiati 2010; Hamman 1983; Swinnen 2010).

Transverse groin incision has the following potential disadvantages over vertical incisions.

• It may have a higher rate of lymphatic complications owing to increased injury to lymph vessels (Roberts 1993).

• Contrasting with the vertical incision, the transverse incision cannot be extended to the inguinal ligament, in order to access the iliac artery, or along the femoral artery.

When considering these controversial points, it is important to establish which surgical technique has a lower rate of complications, greater safety and efficacy. This would reduce surgical morbidity and hospital expenses, ensure early rehabilitation, and also help both physician and patient choose which surgical technique should be used. It would thus become the 'gold standard' surgical technique for surgical access to the femoral artery.

Why it is important to do this review

Complications associated with surgical access to the femoral artery are frequent. They are reported to occur in 2.8% to 44% of cases (Arianne 2009; Van Himbeeck 1992). Soft tissue infection account for 0% to 5.9% of complications (Garner 1986; Chester 1992); endoprosthesis infection account for 0.9% to 6% of complications (Johnson 1988; Szilagy 1972); lymphatic problems are reported in 1% to 5% of cases (Caiati 2010; Shermak 2005; Roberts 1993). Death related to the surgical procedure (which includes incision) occurs in 1.7% of the patients (Holm 1985). Complications are not only responsible for increasing morbidity but are also linked to prolonged hospitalization periods (with extended stays ranging from 2 to 120 days in the case of lymphatic problems [Shermak 2005]); late recovery; additional surgical intervention; and increased hospital expenses (Caiati 2010). Therefore any method leading to a reduced rate of complications from surgical access to the femoral artery is highly welcome.

Some studies have shown that transverse groin incision has a lower rate of complications, especially with regard to infection and lymphatic problems (lymphorrhea/lymphocele), and can be used both in endovascular procedures and in open surgery (Swinnen 2010). Other studies, however, have failed to reveal similar advantages over the classic vertical groin incision technique (Caiati 2010; Roberts 1993).

The aim of this systematic review is to compare the effectiveness and safety of the transverse groin incision approach to the femoral artery with the vertical groin incision approach. Accordingly, we expected the review to provide evidence to support the surgical decision‐making process of physicians and patients, indicating which technique is associated with the best results. In addition, we expected that the outcome of the study would serve as a base to establish the adopted conduct to be standardized and incorporated into a body of consensual guidelines, such as those of the European Society of Vascular Surgery (ESVS), Society of Vascular Surgery (SVS), American Heart Association (AHA) and others.

Objectives

To evaluate the efficacy and safety of transverse groin incision compared with vertical groin incision for accessing the femoral artery in endovascular surgical procedures and open surgery.

Methods

Criteria for considering studies for this review

Types of studies

We included randomized controlled trials (RCTs) and quasi‐randomized trials (qRCTs) that compare transverse and vertical groin incision, during either endovascular or open surgery procedures.

Unlike RCTs, where the allocation method is truly randomized, a qRCT uses a method of allocating participants that is not truly random; for example the use of odd or even hospital number, date of birth or alternation techniques to allocate treatment groups.

Types of participants

We included all patients who were operated on and submitted to femoral artery access, either in open surgery (treatment of thromboembolic disease, bifemoral aortic bypass or infrainguinal bypass), or in endovascular surgery (endovascular repair of abdominal aortic aneurysm (EVAR), thoracic endovascular aneurysm repair (TEVAR), and transcatheter aortic valve implantation (TAVI)). We planned to conduct the analysis in different groups of participants, with or without open wounds in the lower limbs at the time of surgery, including ulcer or gangrene, such as Fontaine IV or Rutherford III (Figure 2).

2.

Fontaine and Rutherford classifications.

Types of interventions

• Transverse versus vertical groin incision in surgical techniques such as:

• • aortic‐femoral, iliac‐femoral, axillary‐femoral bypass;

  • infra‐inguinal, femoro‐femoral bypass;

  • treatment of thromboembolic disease;

  • EVAR, TEVAR or TAVI;

  • hybrid surgery;

  • arterial surgery in general.

We planned to evaluate each subgroup of participants undergoing a given surgery to better evaluate the results and avoid bias. We also planned to separate participants with or without open ulcer or gangrene wounds, such as Fontaine IV or Rutherford III wound types.

Types of outcome measures

Primary outcomes

• Surgical site infection (superficial incisional infection, deep incisional infection or organ/space infection) (≤ 30 days; > 30 days)

The Szilagyi Classification defines surgical site infection as follows.

• • Grade 1: cellulitis involving the wound.

  • Grade 2: infection involving the subcutaneous tissue.

  • Grade 3: infection involving the vascular prosthesis.

Surgical site infection usually manifests within the first 30 days after surgery (Rutherford 2016). It develops with the following signs and symptoms: pain, heat, hyperemia, local hyperthermia, fever and localized abscess formation (with or without spontaneous drainage through the incision). It may be correlated with late graft infection.

• Lymphatic complications (e.g. lymphorrhea/lymphatic leakage, lymphocele) (≤ 30 days; > 30 days)

Complication caused by injury to lymphatic vessels, often after exposure of the femoral artery in patients undergoing infrainguinal revascularization. Observed in 0.8% to 6.4% of cases (Boaventura 2007).

The distinction between lymphocele and lymphatic fistula is important, since the former has low infectious potential, while the latter may serve as a pathway for the infection of prosthetic grafts (Rossi 2014).

• Infection of the vascular graft (endoprosthesis/prosthesis) (early: ≤ 4 months; late: > 4 months)

Early infections (≤ 4 months after graft implantation) are correlated with Szilagyi grade 3 wound infections (infection involving the vascular prosthesis). These infections are caused by virulent bacteria acquired in the hospital and are associated with sepsis (Rutherford 2016).

Late infections are the result of graft colonization by low‐virulence organisms, such as Staphylococcus epidermidis or, less often, by Candida species.

Most early graft infections occur after hospital discharge (usually after 1 to 3 months) and involve extracavitary grafts. Cavitary (i.e. aortic) graft infections manifest as late infections (> 4 months), with an average onset time of more than 40 months.

Early and late infections may be accompanied by total or partial graft involvement, according to the Bunt classification. The categories of this terminology include: peri‐graft infection (total graft: P0, P1; partial graft: P2 and P3); graft‐enteric erosion (GEE); graft‐enteric fistula (GEF); and aortic stump infection.

Secondary outcomes

• Prolonged hospitalization (> 1 week)

• Reoperative surgery (≤ 30 days; > 30 days)

• Death

• Neurological deficit (e.g. paresthesia)

• Amputation

• Graft patency

• Postoperative pain (Numerical Rating Scale for Pain (NRS ‐ Figure 3) (1 day, 1 week, 1 month)

3.

Numerical Rating Scale: participants are asked to circle a number from 0 to 10, 0 to 20 or 0 to 100 that best fits their pain intensity. Zero usually represents 'no pain at all', whereas the upper limit represents 'the worst pain possible' (Haefeli 2006).

Search methods for identification of studies

Electronic searches

The Cochrane Vascular Information Specialist conducted systematic searches of the following databases for randomized controlled trials and controlled clinical trials without language, publication year or publication status restrictions.

• Cochrane Vascular Specialised Register via the Cochrane Register of Studies (CRS Web searched from inception to 17 February 2020);

• Cochrane Central Register of Controlled Trials (CENTRAL) Cochrane Register of Studies Online (CRSO 2020, Issue 1);

• MEDLINE (Ovid MEDLINE® Epub Ahead of Print, In‐Process & Other Non‐Indexed Citations, Ovid MEDLINE® Daily and Ovid MEDLINE®) (searched from 1 January 2017 to 17 February 2020);

• Embase Ovid (searched from 1 January 2017 to 17 February 2020);

• CINAHL EBSCO (searched from 1 January 2017 to 17 February 2020);

• AMED Ovid (searched from 1 January 2017 to 17 February 2020).

The Information Specialist modelled search strategies for other databases on the search strategy designed for CENTRAL. Where appropriate, they were combined with adaptations of the Highly Sensitive Search Strategy designed by Cochrane for identifying randomized controlled trials and controlled clinical trials (as described in the Cochrane Handbook for Systematic Reviews of Interventions Chapter 6, Lefebvre 2011). Search strategies for major databases are provided in Appendix 1.

The Cochrane Vascular Information Specialist searched the following trials registries on 17 February 2020.

• World Health Organization International Clinical Trials Registry Platform (who.int/trialsearch);

• ClinicalTrials.gov.

The authors carried out additional searches in the IBECS databases on 26 March 2020. The search strategy was designed by the review authors and verified by the Cochrane Brazil Information Specialist. See Appendix 2 for details on the search strategy that was used for the review authors' search.

Searching other resources

We checked the bibliographies of included trials for further references to relevant trials. We contacted authors of the included trials for any possible unpublished data.

Data collection and analysis

Selection of studies

We examined the titles and abstracts to select the relevant reports after merging the search results and removing duplicate records. Two review authors (MVCRC, FCN) evaluated the trials independently to determine if they were appropriate for inclusion. We resolved disagreements by discussion within the review team. We then retrieved and examined the full text of the relevant trials for compliance with eligibility criteria. If a trial did not meet the eligibility criteria, we excluded the trial and documented the reason for exclusion.

Data extraction and management

Two review authors (MVCRC, FCN) independently extracted and collected data on paper data‐extraction forms. We resolved any disagreement by discussion within the review team. Where available, we collected the following information.

• Characteristics of the study: details of the publication (e.g. year, country, authors); study methodology (e.g. age, comorbidities, surgery performed, complications presented, period in which complications developed after intervention, treatment performed, evolution); details of the intervention (e.g. surgical reintervention, amputation, time of antibiotic therapy); number of participants randomized in each treatment group; number of participants with complications in each group; number of participants lost to follow‐up; duration of follow‐up; cost of treatment.

• Results: outcomes measured; recorded outcome time points.

Assessment of risk of bias in included studies

Two review authors (MVCRC, FCN) independently assessed the included studies for risk of bias using Cochrane's 'Risk of bias' tool, as described in Section 8.5 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We planned to resolve disagreements by discussion within the review team, if necessary. We assessed the following domains and rated them as having a low, unclear, or high risk of bias.

• Random sequence generation

• Adequate concealment of allocation

• Blinding of participants and personnel

• Blinding of outcome assessment

• Incomplete outcome data

• Selective outcome reporting

• Other potential threats to validity

Blinding of participants or personnel would be difficult—if not impossible—to assess, since both the participant and the team carrying out the procedure would be aware of the performed intervention.

We report the judgement for each individual study in the 'Risk of bias' tables located in the 'Characteristics of included studies' section. We planned to contact the study author(s) to seek clarification in cases of uncertainty regarding data.

Measures of treatment effect

We used risk ratio (RR) to calculate the measure of effect for dichotomous data and mean difference (MD) for continuous data, with the same or standardized mean difference (SMD) for continuous data with different scales, all with 95% confidence intervals.

Unit of analysis issues

For groin‐incision‐related outcomes, such as surgical site infection and lymphatic complications, we considered each groin incision as the unit of analysis. For participant‐related outcomes, such as death, hospitalization and post‐operative pain, we considered the participant as the unit of analysis. Participants who underwent bilateral groin incision, but with different surgeries (e.g. iliac‐femoral shunt on one side and femoro‐popliteal shunt on the other) had groin incision as the unit of analysis.

Dealing with missing data

We included all available data and contacted the study authors to request missing data, where necessary. We contacted study authors on three occasions but unfortunately did not receive a response. We report drop‐out rates in the 'Characteristics of included studies' tables of the review, and we used intention‐to‐treat analysis.

Assessment of heterogeneity

We quantified the inconsistency between the estimates grouped using the I² statistic (where I² = ((Q − df)/Q) × 100% where Q is the statistical Chi², and 'df' represents the degree of freedom).

This illustrates the percentage of variability estimates in the effect resulting from heterogeneity rather than a sampling error (Higgins 2011). We interpreted the thresholds for the I² statistic as follows: 0% to < 30% = low heterogeneity; 30% to < 60% = moderate heterogeneity; 60% to < 90% = substantial heterogeneity; and more than 90% = considerable heterogeneity (Higgins 2011).

Assessment of reporting biases

We planned to evaluate the presence of publication bias and other bias reports using funnel plots, if sufficient studies (more than 10) were identified for inclusion in the meta‐analysis (Higgins 2011).

Data synthesis

We synthesized the data using Review Manager 5 software (Review Manager 2014). We used the fixed‐effect model to synthesize data if there were low levels of heterogeneity. If there was substantial heterogeneity, we used a random‐effects model. If there was considerable heterogeneity we planned not to conduct a meta‐analysis but to describe the data in a narrative form in the text.

Subgroup analysis and investigation of heterogeneity

If sufficient data were available, we planned to perform subgroup analyses for the following factors.

• Type of planned surgery (e.g. aortic‐femoral, iliac‐femoral, axillary‐femoral bypass; infra‐inguinal, femoro‐femoral bypass; treatment of thromboembolic disease; EVAR, TEVAR or TAVI; hybrid surgery; arterial surgery in general).

• Participants with or without open wounds in the lower limbs at the time of surgery, including ulcer or gangrene, such as Fontaine IV or Rutherford III (Figure 2).

• Existence of comorbidities: BMI > 30 kg/m², smoking habit (current or previous), diabetes mellitus, and hypoalbuminemia (< 3.5 mg/dL).

• Participants who underwent previous inguinal surgery vs those who did not, to evaluate whether they are different in terms of complications (lymphatic complications, endoprosthesis infection, vascular graft infection, neurological deficit).

• Surgeon‐related: technique of skin closure (staples, nylon stitches, continuous or interrupted, vs subcutaneous); and technique of subcutaneous tissue closure, running vs interrupted sutures.

• Surgery performed under emergency conditions vs elective surgery, to assess surgical complications (lymphatic complications, infections).

• Procedure‐related: duration of surgery, intra‐ and postoperative glucose control, intraoperative temperature, chronic diabetes control, skin preparation, timing of antibiotic administration.

Sensitivity analysis

We planned to use sensitivity analyses to investigate the robustness of the findings for the primary and secondary outcomes concerning study quality, the source of the data (published or unpublished) and analysis method. We also planned to perform sensitivity analyses by removing studies with high risk of (methodological) bias to assess whether excluding studies with high risk of bias would significantly change the results.

Summary of findings and assessment of the certainty of the evidence

We prepared Table 1 to describe the key information presented in the review comparing vertical versus transverse groin incision for patients who underwent femoral artery access during vascular procedures. We report on the following outcomes: surgical site infection, lymphatic complications, infection of the vascular graft (endoprosthesis/prosthesis), prolonged hospitalization, amputation, and death, as described in the Types of outcome measures section. We based this table on the methods described in Chapters 11 and 12 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We evaluated the certainty of the evidence for each outcome as high, moderate, low, or very low based on criteria of risk of bias, inconsistency, indirectness, imprecision and publication bias using the GRADE approach (Atkins 2004; Guyatt 2008).

Results

Description of studies

See: Characteristics of included studies;Characteristics of excluded studies.

Results of the search

Initially, we retrieved 1873 records through database searching and searching reference lists. After resolving duplicates, we selected 838 records for the screening process. After discarding 829 reports that were clearly not relevant, we identified nine records as potentially eligible studies. After the full‐text assessment, we included two studies in the review. The process of selection of the studies is described in Figure 4.

4.

Study flow diagram.

Included studies

We included one RCT (Chester 1992), and one qRCT in this review (Swinnen 2010).

Chester 1992 conducted a study with 149 participants (167 groins) who underwent inguinal access surgery to access the femoral artery. The study authors compared transverse versus vertical inguinotomies. A total of 85 groin incisions were randomized for vertical access; and 82 for transverse access.

The study was a single‐center trial, conducted in a community hospital. The types of procedures that were undertaken included open vascular surgery, femoropopliteal bypass (reversed saphenous), aortobifemoral bypass, axilobifemoral bypass and femorofemoral crossover graft. The participants included people with severe intermittent claudication, rest pain and gangrene.

Swinnen 2010 carried out a study with 88 patients (116 groins) undergoing inguinal access surgery to access the femoral artery. The study authors compared transverse versus vertical inguinotomies. A total of 61 groin incisions were randomized for vertical access, and 55 for transverse access.

The study was a multi‐center trial in university hospital centers. The types of procedures that were undertaken included suprainguinal surgery (e.g. aortobifemoral graft), inguinal surgery (e.g. transfemoral embolectomy) and infrainguinal surgery (e.g. femoropopliteal bypass). The participants included people with critical limb ischemia and those requiring aortic repair.

The outcome 'wound infection' was assessed in both studies (Chester 1992; Swinnen 2010). Swinnen 2010 also evaluated the following outcomes relevant for this review: lymphatic complications (lymphocele and lymphorrhea). Neither of the studies reported whether the patients were operated on urgently or electively, or whether the patients submitted to surgery had lesions/wounds on their lower limbs at the time of surgery. This information could influence some outcomes, such as infection.

Full details on the individual two studies are available in Characteristics of included studies.

Excluded studies

The specific reasons why the trials were excluded are explained in the Characteristics of excluded studies table. In sum, the studies conducted by Caiati 2010, Parikh 2018, Roberts 1993, and Siracuse 2018 were not randomized controlled trials. The studies conducted by Haaverstad 1995 and Vierhout 2015 did not address the topic of this review, i.e. they did not compare the surgical techniques of transverse versus vertical groin incision.

Risk of bias in included studies

The 'Risk of bias' assessments for all included studies are described in Figure 5 and Figure 6.

5.

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

6.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Allocation

The study conducted by Chester 1992 was a randomized clinical trial. We classified it as having a low risk of bias for this domain. The drawing card method was used for randomization; however, the allocation concealment method was not detailed. We therefore classified allocation concealment as having an unclear risk of bias.

The study conducted by Swinnen 2010 was a quasi‐randomized clinical trial. We classified it as having a high risk of bias for this domain. The last digit of the hospital number was used as the randomization method in this study, according to even or odd numbers. The allocation concealment method used was not detailed. We therefore classified the allocation concealment used as having an unclear risk of bias.

Blinding

In conducting the comparative study of the two surgical techniques, Chester 1992 and Swinnen 2010 failed to perform blinding of the participants and personnel, and also of outcome assessments. We therefore classified them as having a high risk of bias for these domains.

Although the blinding of participants and personnel was not possible in the context of the studies, there was no apparent reason why blinding of outcome assessments was not performed. Blinding of outcome assessments can lower the risk of outcomes being measured in a biased manner, where one of the groups submitted to a particular intervention could be unduly favored (detection bias).

Incomplete outcome data

We classified the Swinnen 2010 study as having a low risk of attrition bias, since the authors reported the results of all the participants and all the participant losses that occurred during the follow‐up period. The intervention groups were similar in regard to participant losses.

We classified the Chester 1992 study as having a high risk of attrition bias. In this study, six participants were lost after randomization. Although the respective causes of the losses were reported, the study authors did not report which groups the lost participants belonged to. This could have influenced the results and the analysis of the outcomes.

Selective reporting

We classified the Chester 1992 and Swinnen 2010 studies as having a low risk of bias as the authors reported on their predefined outcomes. Chester 1992 assessed surgical wound infection; and Swinnen 2010 assessed wound complications (wound infection, wound breakdown and lymphatic leak).

Other potential sources of bias

We classified the studies conducted by Chester 1992 and Swinnen 2010 as having a low risk of bias for this domain.

Effects of interventions

See: Table 1

Chester 1992 and Swinnen 2010 described postoperative outcomes; however, these were analyzed after different follow‐up periods. Chester 1992 evaluated the complications observed on the 10^th postoperative day, whereas Swinnen 2010 reported on the 28^th postoperative day.

Primary outcomes

Surgical site infection (superficial incisional infection, deep incisional infection or organ/space infection) (≤ 30 days; > 30 days)

Regarding surgical wound infection, analysis of the studies conducted by Chester 1992 and Swinnen 2010 revealed a lower rate of complications for transverse groin incision than for vertical groin incision (RR 0.25, 95% CI 0.08 to 0.76; 2 studies; 283 groins). There was low heterogeneity between the studies (I² = 0; Analysis 1.1).

1.1. Analysis.

Comparison 1 Transverse versus vertical groin incision for femoral artery approach, Outcome 1 Wound infection.

Lymphatic complications (e.g. lymphorrhea/lymphatic leakage, lymphocele) (≤ 30 days; > 30 days)

The study performed by Swinnen 2010 found evidence of a difference between transverse and vertical groin incision in regard to the lymphatic complications outcome: lymphorrhea/lymphatic leak (RR 0.46, 95% CI 0.20 to 1.02; 1 study; 116 groins; Analysis 1.2); and lymphocele/lymphatic collection (RR 2.77, 95% CI 0.92 to 8.34; 1 study; 116 groins; Analysis 1.3).

1.2. Analysis.

Comparison 1 Transverse versus vertical groin incision for femoral artery approach, Outcome 2 Lymphatic leak.

1.3. Analysis.

Comparison 1 Transverse versus vertical groin incision for femoral artery approach, Outcome 3 Lymphatic collection.

Infection of the vascular graft (endoprosthesis/prosthesis) (early: ≤ 4 months; late: > 4 months)

This outcome was not evaluated in the two included studies (Chester 1992; Swinnen 2010).

Secondary outcomes

The secondary outcomes—prolonged hospitalization (> 1 week), reoperative surgery (≤ 30 days, > 30 days), death, neurological deficit (e.g. paresthesia), amputation, graft patency, and postoperative pain, amputation—were not evaluated in the two included studies (Chester 1992; Swinnen 2010).

Subgroup and sensitivity analyses

Due to the limited available data, we were unable to perform the planned subgroup and sensitivity analyses.

Discussion

Summary of main results

This review aimed to assess the efficacy and safety of transverse groin incision compared with vertical groin incision to gain access to the femoral artery, in both endovascular surgical procedures and open surgeries. The two studies we included were those conducted by Chester 1992 and Swinnen 2010, for a total of 283 groins. A meta‐analysis of these studies showed low‐certainty evidence that transverse groin incision resulted in a lower risk of surgical wound infection (primary outcome) in the 10‐ to 28‐day period following surgery. No evidence of a difference between the two surgical techniques was observed for the other evaluated outcomes (lymphocele and lymphorrhea). There was low heterogeneity between the studies for the outcome 'wound infection' (see Table 1).

Overall completeness and applicability of evidence

We included two clinical trials (one randomized ‒ Chester 1992; the other quasi‐randomized ‒ Swinnen 2010) to answer the question regarding which surgical technique is safer and more effective to access the femoral artery. The two trials included real‐life populations, used similar selection criteria, and used short clinical follow‐up periods: 10 and 28 days for Chester 1992 and Swinnen 2010, respectively.

Infection of the surgical wound was the only outcome that was assessed in both studies, and therefore the only one that we could submit to a pooled analysis. Evaluation of the infection outcome was, however, performed differently in each study. Chester 1992 considered operative wound infections to be those presenting serosanguinous or purulent discharge. Suture line erythema—a common occurrence—was ignored. Swinnen 2010 classified surgical wound infections according to the criteria established by the Centers for Disease Control and Prevention "Guidelines for the Prevention of Surgical‐Site Infection" (1999). Nevertheless, we observed low heterogeneity (I² = 0) between the studies. The Chi² test result for heterogeneity of the operative wound infection outcome was P = 0.42.

In this systematic review, we found low‐certainty evidence that performing transverse groin incision to access the femoral artery resulted in fewer surgical wound infections compared with performing vertical groin incision. The technique presented a lower risk of surgical wound infection in the 28‐day period following surgery as assessed in two small studies with high risk of bias in critical domains

This result has great relevance for clinical practice, since this surgical procedure is frequently performed in vascular surgeries. A limitation of this systematic review is, however, the small number of studies and small sample sizes. We had expected to find larger numbers of clinical trials and participants in the related literature. Furthermore, we had expected to encounter longer follow‐up periods for the operated patients. These limitations, together with issues surrounding lack of blinding and randomization, contribute to increasing the risk of bias in this systematic review and also to lowering the level of certainty of the evidence provided by it. For this reason, the applicability of the results is limited.

Regarding the lymphatic complications outcome, assessed in one study, we found no evidence of a difference between the two surgical techniques (transverse and vertical groin incision) in the 28‐day period following surgery. It should, however, be taken into account that the clinical follow‐up period was short.

Other planned outcomes—such as infection of the vascular graft, prolonged hospitalization (> 1 week), reoperative surgery (≤ 30 days, > 30 days), death, neurological deficit (e.g. paresthesia), amputation, graft patency, and postoperative pain—were not evaluated in the two included studies.

It is noteworthy that the study by Chester 1992 was published before EVAR or TEVAR were commercially available. Therefore it did not evaluate patients who underwent endovascular procedures, and its primary outcome was surgical wound infection. Although the sample evaluated only included patients who underwent open surgery, we believe that the results pertaining to this outcome would be the same for patients undergoing endovascular surgery (EVAR, TEVAR and TAVI).

Possible confounding factors potentially influencing the outcome of any surgical procedure should also be considered, as follows.

• Surgeon‐related factors, such as experience, time required to perform a given surgical procedure, technical rigor.

• Factors related to pre‐ and postoperative care (such as antibiotic prophylaxis, antisepsis and asepsis, specialized nursing care with dressings).

• Patient‐related factors, such as malnutrition, pre‐existing comorbidities, adherence to medical instructions, personal hygiene.

Quality of the evidence

The certainty of the evidence provided by the studies used in the comparison is summarized in Table 1.

We rated the certainty of the evidence as low for wound infection 10 to 28 days following surgery, and very low for lymphatic complications (namely lymphatic leak and lymphatic collection) 28 days following surgery.

We downgraded the certainty of the evidence for wound infection by one level due to serious limitations in the design (there was a high risk of bias in critical domains) and by a further level due to imprecision, likely caused by the small number of studies and participants.

We downgraded the certainty of evidence for lymphatic complications by one level due to serious limitations in the design (there was a high risk of bias in critical domains), and by two further levels because of imprecision (small number of participants and only one study included).

Potential biases in the review process

The studies used different follow‐up periods for the outcomes evaluated. Chester 1992 followed the patients for 10 days, whereas Swinnen 2010 did so for 28 days. The only outcome that was similar for the two studies, and consequently allowed us to make a combined analysis, was surgical wound infection; the other outcomes evaluated were not similar.

We attempted to contact the authors of the studies to obtain some originally unreported information regarding the following questions: how many patients sustained a surgical wound infection after the 10‐day period (only Swinnen 2010)? Among those patients who had infectious complications, how many were operated on urgently or electively? How many patients presented trophic lesions in the lower limbs? How many patients received antibiotic therapy or antibiotic prophylaxis? Unfortunately, the authors did not respond to the questions raised.

Agreements and disagreements with other studies or reviews

We identified two small studies eligible for this review. Our search did, however, retrieve some non‐randomized studies on the proposed subject of transverse versus vertical groin incision, reporting contrasting results from each other (Caiati 2010; Parikh 2018; Roberts 1993).

Roberts 1993 published a retrospective cohort study comparing 193 patients undergoing 211 surgical procedures, totaling 316 groin wounds. The patients were followed up for six years. The surgeries performed were 91 aortic, 15 extra‐anatomic, and 105 infra‐renal revascularizations. Of these surgeries, 94 involved vertical incisions and 86 involved transverse incisions. The outcomes evaluated were lymphatic complications after each type of surgical access (transverse versus vertical groin incision). The study was not randomized, and the inguinal access procedure was chosen according to the surgeon's preference. Overall, six out of 73 patients undergoing transverse incisions developed lymphocele, versus only one out of 120 patients who underwent vertical incisions (8% versus 1%, P < 0.01). Put in other terms, lymphocele developed in two out of 184 vertical incisions, and in six out of 132 transverse incisions (4% versus 1%, P = 0.06). According to the author, the risk of developing lymphoceles can be virtually annulled by vertical dissection of the femoral vessels, and subsequent meticulously performed closure of the surgical wound.

Caiati 2010 published a prospective cohort study that evaluated patients undergoing endovascular aortic aneurysm correction using femoral surgical access via transverse groin incision. The authors' hypothesis was that transverse groin incision would be associated with a lower rate of surgical complications than vertical groin incision. The outcomes evaluated were cellulitis, subcutaneous purulence, femoro‐femoral graft infection, lymphocele, and lymphocutaneous fistula. The study evaluated 100 consecutive patients from June 1998 to May 1999. Twenty‐four patients received an aortic‐aortic endoprosthesis; 33 patients, a bifurcated endoprosthesis; and 41 patients, uni‐iliac aortic endoprostheses with revascularization of the femoro‐femoral artery. There was a total of 176 groin lesions. Twelve patients were women (12%) and 88 were men (88%). The mean age was 75.3 years. Infectious and lymphatic surgical wound complications occurred in five out of 176 groin lesions (2.8% incidence). All of the surgical wound complications were grade I, according to the classification by Szilagy 1972, and consisted of cellulitis (1); lymphocele (1); or lymphatic fistula (3). According to the study authors, transverse inguinal access to the femoral artery "provides ample exposure for these procedures with decreased postoperative wound morbidity." The study authors suggest that this technique should be the preferred technique for endovascular procedures, and that it may also be useful for patients undergoing lower limb arterial bypass procedures.

Parikh 2018 published a prospective cohort study comparing 150 patients undergoing 156 lower limb revascularizations (aortobifemoral bypass and axillary bifemoral bypass, iliofemoral endarterectomy, deep endarterectomy, iliofemoral bypass for distal deviation, and iliac artery aneurysm repair). Of these revascularizations, 85 involved transverse incisions (54%) and 71 involved vertical incisions (46%). The primary outcome evaluated was complication of the surgical wound, including infection, lymphocele, hematoma, suture dehiscence, pseudoaneurysm, and necrosis. The other outcomes studied were the need for reoperation to treat surgical wound complications, wound vacuum therapy, and the need for flap surgery for surgical wound closure. The mean overall follow‐up period was 220 days. Overall, patients undergoing transverse groin incision had significantly fewer surgical wound complications than those submitted to vertical groin incision, with overall rates of 7% and 42%, respectively (P < 0.001). Furthermore, transverse incisions were associated with lower rates of reoperation (5% versus 23%, P < 0.001), need for vacuum therapy (6% versus 15%, P < 0.05), and muscle flap closure (0% versus 13%, P < 0.001).

Authors' conclusions

Implications for practice.

Evidence of low certainty suggests that surgical wound infection in the 28‐day period following surgery occurs less frequently in transverse groin incision than in vertical groin incision to access the femoral artery. Evidence of very low certainty indicated that there was no evidence of a difference between the two surgical techniques to access the femoral artery with respect to lymphocele and lymphorrhea in the first 28 days following surgery.

Limitations of this systematic review are limited availability of clinical trials, small number of participants, short clinical follow‐up periods and high risk of bias in critical domains. For this reason, the applicability of the results is limited. New studies may find contrasting results or impart greater consistency to this systematic review.

High‐quality studies are needed to enable a comparison of the two surgical techniques with respect to other outcomes, such as infection of the vascular graft (endoprosthesis/prosthesis), neurological deficit (e.g. paresthesia), graft patency, postoperative pain, amputation, prolonged hospitalization, reoperative surgery, and death.

Implications for research.

We suggest that future comparative studies between the transverse and vertical groin incision techniques:

• clearly describe the methods of randomization and allocation concealment;

• include a greater number of participants, since these are surgical techniques regularly performed in vascular surgeries;

• perform a clinical follow‐up of patients during longer periods, reporting on the need for reoperation, graft patency, hospitalization time, postoperative pain, and progression to limb amputation or death;

• use previously established criteria to define certain outcomes in an objective and reproducible manner (such as the Szilagyi classification for surgical site infection);

• perform blinding of the outcome assessments (detection bias), e.g. by using photos that make it impossible for the evaluator to tell which surgical technique was used;

• provide more detailed information about patients and surgical procedures in order to avoid possible bias, including:

  • nature of the surgery (emergency or elective);

  • presence or absence of lesions (infected or otherwise) in the lower limbs at the time of surgery;

  • history of previous inguinal surgery;

  • duration of surgery;

• build a flowchart with a detailed description of the participant losses throughout the study;

• follow the CONSORT guidelines.

Notes

Parts of the Methods section of this review are based on a standard template established by Cochrane Vascular.

Appendices

Appendix 1. Database searches 17 February 2020

Source	Search strategy	Hits retrieved
CENTRAL	#1 MESH DESCRIPTOR Femoral Artery EXPLODE ALL TREES 909 #2 femoral*:TI,AB,KY 9617 #3 #1 OR #2 9617 #4 MESH DESCRIPTOR GROIN EXPLODE ALL TREES 119 #5 Infrainguinal*:TI,AB,KY 169 #6 Inguinotomy:TI,AB,KY 0 #7 inguinofemoral:TI,AB,KY 13 #8 Inguinal*:TI,AB,KY 2841 #9 Groin*:TI,AB,KY 955 #10 #4 OR #5 OR #6 OR #7 OR #8 OR #9 3436 #11 #3 AND #10 291	06.11.2018: 291 17.02.2020: 165
ClinicalTrials.gov	Femoral | GROIN OR Infrainguinal OR Inguinotomy OR inguinofemoral OR Inguinal	06.11.2018: 77 17.02.2020: 10
ICTRP Search Portal	Femoral | GROIN OR Infrainguinal OR Inguinotomy OR inguinofemoral OR Inguinal	06.11.2018: 22 17.02.2020: 8
MEDLINE (Ovid MEDLINE® Epub Ahead of Print, In‐Process & Other Non‐Indexed Citations, Ovid MEDLINE® Daily and Ovid MEDLINE®) 1946 to present (2017 onwards)	1 exp Femoral Artery/ 2 femoral*.ti,ab. 3 1 or 2 4 exp GROIN/ 5 Inguinal*.ti,ab. 6 Groin*.ti,ab. 7 Infrainguinal*.ti,ab. 8 Inguinotomy.ti,ab. 9 inguinofemoral.ti,ab. 10 or/4‐9 11 3 and 10 12 randomized controlled trial.pt. 13 controlled clinical trial.pt. 14 randomized.ab. 15 placebo.ab. 16 drug therapy.fs. 17 randomly.ab. 18 trial.ab. 19 groups.ab. 20 or/12‐19 21 exp animals/ not humans.sh. 22 20 not 21 23 11 and 22	06.11.2018: 67 17.02.2020: 68
Embase 1974 to present (2017 onwards)	1 exp femoral artery/ 2 femoral*.ti,ab. 3 1 or 2 4 exp inguinal region/ 5 Inguinal*.ti,ab. 6 Groin*.ti,ab. 7 Infrainguinal*.ti,ab. 8 Inguinotomy.ti,ab. 9 inguinofemoral.ti,ab. 10 or/4‐9 11 3 and 10 12 randomized controlled trial/ 13 controlled clinical trial/ 14 random$.ti,ab. 15 randomization/ 16 intermethod comparison/ 17 placebo.ti,ab. 18 (compare or compared or comparison).ti. 19 ((evaluated or evaluate or evaluating or assessed or assess) and (compare or compared or comparing or comparison)).ab. 20 (open adj label).ti,ab. 21 ((double or single or doubly or singly) adj (blind or blinded or blindly)).ti,ab. 22 double blind procedure/ 23 parallel group$1.ti,ab. 24 (crossover or cross over).ti,ab. 25 ((assign$ or match or matched or allocation) adj5 (alternate or group$1 or intervention$1 or patient$1 or subject$1 or participant$1)).ti,ab. 26 (assigned or allocated).ti,ab. 27 (controlled adj7 (study or design or trial)).ti,ab. 28 (volunteer or volunteers).ti,ab. 29 trial.ti. 30 or/12‐29 31 11 and 30 32 (2017* or 2018*).dc. 33 31 and 32	06.11.2018: 183 17.02.2020: 189
CINAHL (2017 onwards)	S28 S26 AND S27 S27 EM 2017 OR EM 2018 S26 S11 AND S25 S25 S12 OR S13 OR S14 OR S15 OR S16 OR S17 OR S18 OR S19 OR S20 OR S21 OR S22 OR S23 OR S24 S24 MH "Triple‐Blind Studies" S23 MH "Double‐Blind Studies" S22 MH "Single‐Blind Studies" S21 MH "Crossover Design" S20 MH "Factorial Design" S19 MH "Placebos" S18 MH "Clinical Trials" S17 TX "multi‐centre study" OR "multi‐center study" OR "multicentre study" OR "multicenter study" OR "multi‐site study" S16 TX crossover OR "cross‐over" S15 AB placebo* S14 TX random* S13 TX trial* S12 TX "latin square" S11 S3 AND S10 S10 S4 OR S5 OR S6 OR S7 OR S8 OR S9 S9 TX inguinofemoral S8 TX Inguinotomy S7 TX Infrainguinal* S6 TX Inguinal* S5 TX GROIN S4 (MH "Groin") S3 S1 OR S2 S2 TX femoral* S1 (MH "Femoral Artery")	06.11.2018: 14 17.02.2020: 19
Total before deduplication		06.11.2018: 802 17.02.2020: 468
Total after deduplication		06.11.2018: 696 17.02.2020: 378
2018 and 2020 searches combined and deduplicated		825

Appendix 2. Authors' IBECS search strategy (26 March 2020)

Additional search strategy: "P" AND "I"

P:

Mh:"Arteria Femoral" OR Mh:"Femoral Artery" OR Mh:"Artéria Femoral" OR (Arteria Femoral) OR (Femoral Artery) OR (Artéria Femoral) OR (Arteries, Femoral) OR (Artery, Femoral) OR (Femoral Arteries) OR mh:A07.231.114.351$ OR Mh:"Peripheral Vascular Diseases" OR Mh:"Enfermedades Vasculares Periféricas" OR Mh:"Doenças Vasculares Periféricas" OR (Peripheral Vascular Diseases) OR (Enfermedades Vasculares Periféricas) OR (Doenças Vasculares Periféricas) OR (Angiopatias Periféricas) OR (Doenças Arteriais Periféricas) OR (Diseases, Peripheral Vascular) OR (Angiopathies, Peripheral) OR (Angiopathy, Peripheral) OR (Disease, Peripheral Vascular) OR (Peripheral Angiopathy) OR (Peripheral Vascular Disease) OR (Vascular Disease, Peripheral) OR (Peripheral Angiopathies) OR (Vascular Diseases, Peripheral) OR (Angiopatías Periféricas) OR (Enfermedades Arteriales Periféricas) OR mh:C14.907.617$

AND

I:

Mh:"Procedimientos Quirúrgicos Vasculares" OR Mh:"Vascular Surgical Procedures" OR Mh:"Procedimentos Cirúrgicos Vasculares" OR (Procedimientos Quirúrgicos Vasculares) OR (Vascular Surgical Procedures) OR (Procedimentos Cirúrgicos Vasculares) OR (Procedure, Vascular Surgical) OR (Procedures, Vascular Surgical) OR (Surgical Procedure, Vascular) OR (Surgical Procedures, Vascular) OR (Vascular Surgical Procedure) OR (Cirurgia Vascular) OR (groin incision) OR (incisão inguinal) OR inguinotomia OR inguinotomy OR mh:E04.100.814$ OR Mh:"Procedimientos Endovasculares" OR Mh:"Endovascular Procedures" OR Mh:"Procedimentos Endovasculares" OR (Procedimientos Endovasculares) OR (Endovascular Procedures) OR (Procedimentos Endovasculares) OR (Endovascular Techniques) OR (Intravascular Procedures) OR (Intravascular Techniques) OR (Endovascular Procedure) OR (Endovascular Technique) OR (Intravascular Procedure) OR (Intravascular Technique) OR (Procedure, Endovascular) OR (Procedure, Intravascular) OR (Procedures, Endovascular) OR (Procedures, Intravascular) OR (Technique, Endovascular) OR (Technique, Intravascular) OR (Techniques, Endovascular) OR (Techniques, Intravascular) OR (vascular reconstruction) OR (reconstrução vascular) OR E04.100.814.529$ OR E04.502.382$

Appendix 3. Records identified by review authors (additional papers from searching reference lists)

Beirne 2008

Chester 1992

Roberts 1993

Appendix 4. Glossary of terms

Arterial graft: vascular bypass (or vascular graft) is a way of redirecting blood flow via a surgical procedure. It involves using an arterial graft that re‐establishes the blood flow, creating a bypass that diverts blood from a diseased segment, connecting an area with normal blood flow to another relatively normal area. The nomenclature of an arterial graft involves the donor artery and the recipient artery (e.g. aortic shunt to the femoral artery—aortofemoral shunt).

Types of vascular grafts: 1) synthetic grafts (polytetrafluoroethylene graft (Teflon), polyethylene terephthalate (Dacron)); 2) vein allografts from a different person; 3) autografts (veins or arteries from the patient himself/herself).

Cavitary vascular grafts: are derivations performed with grafts (autologous, heterologous or synthetic) that extend beyond the coelomic cavity (peritoneal cavity, retroperitoneum, thoracic cavity). That is, they go beyond the abdominal or thoracic cavity and connect central circulation vessels to other vessels. (e.g. aortic‐femoral graft, iliac‐femoral graft)

Endovascular repair of abdominal aortic aneurysm (EVAR): is the placement of an endoprosthesis, often through the femoral arteries, that involves an inner lining of the abdominal aorta, in order to isolate the diseased segment and prevent lesion rupture. Sealing of the device against the aortic wall is achieved above (proximally) and below (distally) the diseased artery (aneurysm, aortic ulcer, dissection of the arterial wall, or point of vascular trauma).

Endovascular surgery: a surgical procedure in which the surgeon performs remote treatment of circulatory diseases. It encompasses vascular catheterization techniques, aimed at treating diseases of the veins and arteries of the whole body in a minimally invasive manner. The diseases treated are mainly: aortic aneurysms, obstructions of the blood flow to the legs and the brain (extremity and carotid atherosclerotic disease), among others.

Extracavitary vascular grafts: are shunts performed with grafts (autologous, heterologous or synthetic) located in peripheral vessels, such as in upper or lower limbs. They do not extend beyond the coelomic cavity (peritoneal cavity, retroperitoneum, thoracic cavity).

Graft patency: maintenance of blood flow through the graft. It can be subdivided into:

• primary patency: characterized by uninterrupted patency, without the need for new interventions, reflecting the durability of the successful outcome of the procedure;

• primary assisted patency: patency of a blood vessel after endovascular intervention (angioplasty) that treats a stenosing lesion, i.e. one that decreases blood flow, but does not occlude the vessel. It reflects the impact of surveillance on the primary treatment performed;

• Secondary patency: the patency of the graft after an endovascular intervention (angioplasty) that treats occlusion of a previously treated site.

Lymphatic leakage: trauma to the lymphatic system during surgical procedures may result in postoperative lymph leakage (Lv 2017).

Lymphatic fistula:

Lymphatic fistula includes lymphocutaneous fistula and lymphoperitoneal fistula.

• Lymphocutaneous fistula causes the formation of lymphorrhea and other associated lymphatic leakages.

• Lymphoperitoneal fistula is one of the causes of lymphatic ascites and lymphocele.

It frequently occurs after surgical procedures that involve dissection of the axillary and inguinal region, such as surgery for the treatment of urologic, gynecological, and dermatological cancers, as well as for arterial reconstruction.

• Lymphorrhea: is lymphatic exudation in the wound after trauma to the lymphatic vessels. When caused by injury to deep lymph vessels, it may resolve spontaneously or develop into lymphatic ascites or lymphocele (light‐colored fluid, very soggy dressing).

• Lymphocele (lymphocyst): is a cyst filled with clear lymphatic fluid, with no inflammatory or granulomatous reaction at the site of the leakage. It may manifest as bulging near the operative wound. It often occurs within 3 to 8 weeks, or as long as 1 year, after surgery. Most postoperative lymphoceles are asymptomatic, have a self‐limited evolution, and often go undiagnosed and heal spontaneously. Only 4% to 7% of postoperative lymphoceles are symptomatic, manifesting with local pain or discomfort. The average diameter of a symptomatic lymphocele is generally greater than 5 cm. Surgical intervention is proposed when there is pain, infection, lymphorrhea or compression of vital structures. Inguinal lymphocele is also a common complication in surgery that involves the dissection of inguinal lymph nodes. Body mass index (BMI) and the number of resected lymph nodes are predisposing factors.

Many types of postoperative lymphatic leakage have been reported, including lymphatic ascites, lymphocele, lymphorrhea, lymphatic fistula, and some special forms of lymphatic infiltration: chylous ascites (chylous peritoneum), chylous retroperitoneum, chylothorax (Inoue 2016).

Thoracic endovascular aneurysm repair (TEVAR): is an endovascular surgery performed for the treatment of aortic diseases, most commonly abdominal aortic aneurysm (AAA). It involves the placement of an expandable endoprosthesis within the aorta, lining the aorta internally, to redirect the blood flow and protect the aorta from rupture.

Thromboembolic disease: occurs owing to the formation of a clot (thrombus), most often of cardiogenic origin, which is released into the blood stream and occludes a distant blood vessel. The clot can obstruct a vessel in the lungs (pulmonary embolism), brain (cerebrovascular accident), gastrointestinal tract (mesenteric ischemia), kidneys (renal ischemia), or leg (acute arterial occlusion). Thromboembolism is a significant cause of morbidity (disease) and mortality (death), especially in adults and hospitalized patients.

Transcatheter aortic valve implantation (TAVI): a minimally invasive surgical procedure that repairs the aortic valve without removing the old and damaged valve. It fits a replacement valve on the aortic valve. The surgery can be called transcatheter aortic valve replacement (TAVR) or transcatheter aortic valve implantation (TAVI).

Data and analyses

Comparison 1. Transverse versus vertical groin incision for femoral artery approach.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Wound infection	2	283	Risk Ratio (M‐H, Fixed, 95% CI)	0.25 [0.08, 0.76]
2 Lymphatic leak	1		Risk Ratio (M‐H, Fixed, 95% CI)	Totals not selected
3 Lymphatic collection	1		Risk Ratio (M‐H, Fixed, 95% CI)	Totals not selected

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Chester 1992.

Methods	Randomized controlled trial. Patients were randomized by drawing cards, producing equivalent groups with respect to age, sex, smoking, history, diabetes mellitus, indication for surgery and operation planned. When patients required bilateral groin explorations, each wound was treated separately for the purposes of the study.
Participants	149 patients (167 groins) undergoing vascular reconstruction through a groin incision over a period of 30 months were studied. A total of 85 groin incisions were randomized for vertical access; and 82 for transverse access. Those with previous groin incisions were excluded, and where an incision was necessary in both groins, each wound was studied separately. Patients who had undergone previous vascular surgery through a groin incision were excluded. The participants included people with severe intermittent claudication, rest pain and gangrene. The types of procedures that were undertaken included open vascular surgery, femoropopliteal bypass (reversed saphenous), aortobifemoral bypass, axilobifemoral bypass and femorofemoral cross‐over graft.
Interventions	Oblique/transverse versus vertical/longitudinal incisions for access to the femoral vessels for vascular reconstruction.
Outcomes	Participants were followed up for 10 days after the procedure and examined for wound infection. Groin wounds were reviewed by an independent observer at 4, 7 and 10 days postoperatively, and classified into 2 groups: (1) clean or (2) infection with serosanguinous or purulent discharge.
Notes	The study did not report whether the patients were operated on urgently or electively, or whether the patients submitted to surgery had lesions/wounds on their lower limbs at the time of surgery.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	The method of randomization used presents a low risk of bias: "Patients were randomized by drawing cards, producing equivalent groups with respect to age, sex, smoking history, diabetes mellitus, indication for surgery and operation planned."
Allocation concealment (selection bias)	Unclear risk	The method of concealment was not described in sufficient detail to allow a final judgment.
Blinding of participants and personnel (performance bias) All outcomes	High risk	There was no blinding of participants and personnel, and this may have interfered in the outcomes.
Blinding of outcome assessment (detection bias) All outcomes	High risk	There was no blinding of the outcome assessors, and this may have interfered in the outcomes.
Incomplete outcome data (attrition bias) All outcomes	High risk	6 participants were lost after randomization. Although the respective cause of the loss was reported, the study authors did not report what groups the lost participants belonged to. This could have influenced the results and the analysis of the outcomes. "Six patients died within 10 days of surgery, four from myocardial infarcts, and two from cerebral hemorrhage. All had clean wounds up to their death, and all six were excluded from the study."
Selective reporting (reporting bias)	Low risk	The authors clearly stated their predefined outcomes; therefore, we classified their studies as having a low risk of bias.
Other bias	Low risk	The study appears to be free from other sources of bias.

Swinnen 2010.

Methods	Quasi‐randomized trial. Patients undergoing vascular surgery requiring exposure of the femoral artery at 2 institutions (Westmead and Liverpool Hospitals) were randomized to either a vertical or a transverse incision. Patients were excluded from the study if they had undergone previous vascular surgery in the index groin or died before the completion of the study period. Patients were randomized to their incision based on the last digit of the hospital number either even or odd. For bilateral groin incisions, based on the randomization, both groins underwent the same incision. The types of procedures that were undertaken included suprainguinal surgery (e.g. aortobifemoral graft), inguinal surgery (e.g. transfemoral embolectomy) and infrainguinal surgery (e.g. femoropopliteal bypass). The participants included people with critical limb ischemia and aortic repair.
Participants	88 patients (116 groins) were randomized to either vertical (61 inguinotomies) or transverse (55 inguinotomies) incision. One additional patient (one groin) was randomized but excluded from final analysis due to death at day 4 – it is unclear which group this patient was randomized to and no data are presented for this patient. Patients were excluded from the study if they had undergone previous vascular surgery in the index groin or died before the completion of the study.
Interventions	All patients undergoing vascular procedure requiring access to femoral vessels were randomized to either a vertical or transverse incision.
Outcomes	Participants were followed up for 28 days after the procedure and examined for wound infection, wound breakdown, development of lymphatic leak and lymphatic collection. The groin wounds were examined at 4, 10, and 28 days.
Notes	The study did not report whether the participants were operated on urgently or electively, or whether the participants submitted to surgery had lesions/wounds on their lower limbs at the time of surgery.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	The method of randomization used presents a high risk of bias: "Patients were randomized to their incision based on the last digit of the hospital number either even or odd."
Allocation concealment (selection bias)	Unclear risk	The method of concealment was not described in sufficient detail to allow a final judgment.
Blinding of participants and personnel (performance bias) All outcomes	High risk	There was no masking of the participants or personnel. This may have interfered in the outcomes.
Blinding of outcome assessment (detection bias) All outcomes	High risk	There was no masking of the outcome assessors. This may have interfered in the outcomes.
Incomplete outcome data (attrition bias) All outcomes	Low risk	The loss of data was not enough to influence the results. An explanation was given for the participants who were excluded from the study: "One patient (two groins) was excluded before randomization due to the surgeon's request, and one patient (one groin) was excluded from the final analysis due to death at day 4."
Selective reporting (reporting bias)	Low risk	The authors clearly stated their predefined outcomes; therefore, we classified their study as having a low risk of bias.
Other bias	Low risk	The study appears to be free from other sources of bias.

Characteristics of excluded studies [ordered by study ID]

Study	Reason for exclusion
Caiati 2010	Wrong study design: not an RCT
Haaverstad 1995	Wrong comparator: comparison of femoropopliteal bypass reconstruction with either a lateral groin incision versus direct incision over the femoral vessels
Parikh 2018	Wrong study design: not an RCT
Roberts 1993	Wrong study design: not an RCT
Siracuse 2018	Wrong study design: not an RCT and wrong comparator: assessment of VQI registry (a prospective, national database) to compare percutaneous femoral access in endovascular repair versus open femoral access
Vierhout 2015	Wrong comparator: comparison of percutaneous femoral access in endovascular repair versus open femoral access

RCT: randomized controlled trial

Differences between protocol and review

For groin‐incision‐related outcomes, we revised the unit of analysis from participant to groin incision: for groin‐incision‐related outcomes such as surgical site infection and lymphatic complications we considered each groin incision as the unit of analysis. For participant‐related outcomes such as death, hospitalization and post‐operative pain, we considered the participant as the unit of analysis.

We have rephrased inguinotomy to groin incision and longitudinal to vertical to improve the readability and understanding of the review.

Contributions of authors

MVCRC: protocol drafting, acquiring trial reports, trial selection, data extraction, data analysis, data interpretation, review drafting, and future review updates, guarantor of the review.
 JCCB: protocol drafting, trial selection, data interpretation, review drafting, and future review updates.
 FCN: trial selection, data extraction, data analysis, data interpretation, review drafting, and future review updates.
 DGC: protocol drafting, acquiring trial reports, trial selection, data extraction, data analysis, data interpretation, review drafting, and future review updates, guarantor of the review.

Sources of support

Internal sources

• No sources of support supplied

External sources

• Chief Scientist Office, Scottish Government Health Directorates, The Scottish Government, UK.

  The Cochrane Vascular editorial base is supported by the Chief Scientist Office.

Declarations of interest

MVCRC: none known
 JCCB: none known
 FCN: none known
 DGC: none known

New