Advertisement

Editor's Choice – Regional Versus General Anaesthesia in Peripheral Vascular Surgery: a Propensity Score Matched Nationwide Cohort Study of 17 359 Procedures in Denmark

  • Jannie Bisgaard
    Correspondence
    Corresponding author. Department of Anaesthesiology and Intensive Care, Aalborg University Hospital, Hobrovej 18–22, 9000, Aalborg, Denmark.
    Affiliations
    Department of Anaesthesiology and Intensive Care, Aalborg University Hospital, Aalborg, Denmark

    Department of Clinical Medicine, Aalborg University, Aalborg, Denmark
    Search for articles by this author
  • Author Footnotes
    § Present address. Department of Cardiology and Clinical Research, Nordsjaellands Hospital Hillerød, Denmark.
    Christian Torp-Pedersen
    Footnotes
    § Present address. Department of Cardiology and Clinical Research, Nordsjaellands Hospital Hillerød, Denmark.
    Affiliations
    Department of Cardiology, Aalborg University Hospital, Aalborg, Denmark
    Search for articles by this author
  • Bodil S. Rasmussen
    Affiliations
    Department of Anaesthesiology and Intensive Care, Aalborg University Hospital, Aalborg, Denmark

    Department of Clinical Medicine, Aalborg University, Aalborg, Denmark
    Search for articles by this author
  • Kim C. Houlind
    Affiliations
    Department of Vascular Surgery, Lillebælt Hospital, Kolding, Denmark

    Department of Regional Health Research, University of Southern Denmark, Kolding, Denmark
    Search for articles by this author
  • Signe J. Riddersholm
    Affiliations
    Department of Internal Medicine, Randers Regional Hospital, Randers, Denmark
    Search for articles by this author
  • Author Footnotes
    § Present address. Department of Cardiology and Clinical Research, Nordsjaellands Hospital Hillerød, Denmark.
Open ArchivePublished:December 21, 2020DOI:https://doi.org/10.1016/j.ejvs.2020.11.025

      Objective

      Cardiopulmonary comorbidity is common in vascular surgery. General anaesthesia (GA) may impair perfusion and induce respiratory depression. Regional anaesthesia (RA), including neuraxial or peripheral nerve blocks, may therefore be associated with a better outcome.

      Methods

      This was a nationwide retrospective cohort study. All open inguinal and infra-inguinal arterial surgical reconstructions from 2005 to 2017 were included. Data were extracted from national registries. Multivariable linear and logistic regression models and propensity score matching were used. The propensity score was derived by developing a model that predicted the probability that a given patient would receive GA based on age, comorbidity, anticoagulant medication, procedure type, and the urgency of surgery. Matching was performed in four groups based on American Society of Anesthesiologists’ score I – II, score III – V, and gender. Outcome parameters included surgical and general complications (bleeding, thrombosis/embolus, cardiac, pulmonary, renal, cerebral, and >3 days intensive care therapy), length of stay, and 30 day mortality, hypothesising a better outcome after RA.

      Results

      There were 10 509 procedures in the GA group and 6 850 in the RA group. After propensity score matching, 6 267 procedures were included in each group. Surgical and general complications were significantly more common after GA in both matched (3.8 vs. 2.5%, p < .001 and 6.5 vs. 4.2%, p < .001) and unmatched analyses (3.8 vs. 2.5%, p < .001 and 6.5 vs. 4.2%, p < .001). The 30 day mortality rate was significantly higher after GA, in matched and un matched analyses (3.1 vs. 2.4%, p = .019 and 4.1 vs. 2.4%, p < .001). There was no difference in length of stay.

      Conclusion

      RA may be associated with a better outcome, compared with GA, after open inguinal and infra-inguinal peripheral vascular surgery. In the clinical context when RA is not feasible, GA can still be considered safe.

      Keywords

      This study draws attention to the importance of considering regional anaesthesia in patients for peripheral vascular surgery to improve clinical results. Results from this nationwide study indicate a better outcome when using neuraxial anaesthesia and/or peripheral nerve blocks, compared with general anaesthesia, in lower limb arterial revascularisation surgery. This observational trial suggests a significant benefit of regional anaesthesia and underlines the importance of performing a future large scale randomised clinical trial to evaluate this observed treatment benefit.

      Introduction

      Atherosclerosis is a systemic disease affecting multiple organ systems, including cardiac, cerebral and renal function. Atherosclerosis of the lower limbs may result in critical ischaemia and, in the worst cases, gangrene.
      • Nowygrod R.
      • Egorova N.
      • Greco G.
      • Anderson P.
      • Gelijns A.
      • Moskowitz A.
      • et al.
      Trends, complications, and mortality in peripheral vascular surgery.
      ,
      • Nehler M.R.
      • Duval S.
      • Diao L.
      • Annex B.H.
      • Hiatt W.R.
      • Rogers K.
      • et al.
      Epidemiology of peripheral arterial disease and critical limb ischemia in an insured national population.
      Surgical arterial revascularisation is considered a high risk procedure because of the comorbidity in this patient group and the nature of the disease, involving ischaemia reperfusion and potential massive blood loss.
      • Fleisher L.A.
      • Fleischmann K.E.
      • Auerbach A.D.
      • Barnason S.A.
      • Beckman J.A.
      • Bozkurt B.
      • et al.
      2014 ACC/AHA guideline on perioperative cardiovascular evaluation and management of patients undergoing noncardiac surgery: executive summary.
      Planning anaesthesia and surgery for open vascular procedures of the lower limb requires thorough assessment of the patient's physical reserve and ideally involves cardiac and pulmonary evaluation, and perhaps optimisation.
      • Neves S.E.
      Anesthesia for patients with peripheral vascular disease and cardiac dysfunction.
      In many cases, however, the indication is critical ischaemia, and surgery is limb saving, leaving only restricted time for pre-optimisation of the patient. General anaesthesia (GA) has potential disadvantages in patients with cardiopulmonary comorbidity and advanced age.
      • Johnson R.L.
      • Kopp S.L.
      • Burkle C.M.
      • Duncan C.M.
      • Jacob A.K.
      • Erwin P.J.
      • et al.
      Neuraxial vs general anaesthesia for total hip and total knee arthroplasty: a systematic review of comparative-effectiveness research.
      ,
      • Memtsoudis S.G.
      • Rasul R.
      • Suzuki S.
      • Poeran J.
      • Danninger T.
      • Wu C.
      • et al.
      Does the impact of the type of anesthesia on outcomes differ by patient age and comorbidity burden?.
      Positive pressure ventilation affects haemodynamics and pulmonary physiology, which may result in atelectasis and perfusion/ventilation mismatch and GA may affect blood pressure and microcirculation negatively. Decreased tissue perfusion may potentially cause regional ischaemia, which can result in complications, ranging from impaired wound healing to thrombotic events and graft failure and, in the worst cases, organ dysfunction.
      • Turek Z.
      • Sykora R.
      • Matejovic M.
      • Cerny V.
      Anesthesia and the microcirculation.
      Regional anaesthesia (RA) includes neuraxial blockades and peripheral nerve blocks, which include spinal and/or epidural, or peripheral perineural instillation of local anaesthetics, respectively. To date, the evidence for favouring neuraxial anaesthesia over GA is inconclusive.
      • Barbosa F.T.
      • Jucá M.J.
      • Castro A.A.
      • Cavalcante J.C.
      Neuraxial anaesthesia for lower-limb revascularization.
      ,
      • Sgroi M.D.
      • McFarland G.
      • Mell M.W.
      Utilization of regional versus general anesthesia and its impact on lower extremity bypass outcomes.
      In patients using anticoagulants (e.g. because of dysrhythmias, thrombosis or mechanical heart valves), neuraxial blockades are contraindicated.
      • Gogarten W.
      • Vandermeulen E.
      • Van Aken H.
      • Kozek S.
      • Llau J.V.
      • Samama C.M.
      Regional anaesthesia and antithrombotic agents: recommendations of the European Society of Anaesthesiology.
      As an alternative, peripheral nerve blocks are increasingly used in vascular surgery.
      • Yazigi A.
      • Madi-Gebara S.
      • Haddad F.
      • Hayeck G.
      • Tabet G.
      Combined sciatic and femoral nerve blocks for infrainguinal arterial bypass surgery: a case series.
      • Guimarães J. de F.
      • Angonese C.F.
      • Gomes R.K.
      • Junior V.M.
      • Farias C.
      Anesthesia for lower extremity vascular bypass with peripheral nerve block.
      • Selvi O.
      • Bayserke O.
      • Tulgar S.
      Use of femoral and sciatic nerve block combination in severe emphysematous lung disease for femoral popliteal arterial bypass surgery.
      The aim of this retrospective nationwide registry study was to evaluate the effect of GA and RA on outcome after open inguinal and infra-inguinal arterial reconstruction surgery. It was hypothesised that the rates of surgical and general complications (bleeding, thrombosis/embolus, cardiac, pulmonary, renal, cerebral, and >3 days intensive care therapy), length of stay, and 30 day mortality would be lower after RA.

      Methods

      The study was approved by the Danish Data Protection Agency 2008-58-0028, ID 2018–13. In Denmark, registry studies do not need ethical approval.
      This was a retrospective, propensity score matched, cohort study, providing a descriptive analysis of associations between anaesthesia method and outcomes. The cohort comprised all first time open inguinal and infra-inguinal arterial reconstruction procedures in Denmark during the calendar years 2005–2017. All data were registered by the vascular surgeon in charge of the patient and reported to the Danish Vascular Registry. All procedures were included where the question of GA vs. RA was relevant (open arterial revascularisations where RA alone would be sufficient for surgery). Covariates were selected from a clinical perspective, including variables previously defined as risk factors.
      • Kehlet M.
      • Jensen L.P.
      • Schroeder T.V.
      Risk factors for complications after peripheral vascular surgery in 3,202 patient procedures.
      ,
      • Singh N.
      • Sidawy A.N.
      • Dezee K.
      • Neville R.F.
      • Weiswasser J.
      • Arora S.
      • et al.
      The effects of the type of anesthesia on outcomes of lower extremity infrainguinal bypass.
      All residents in Denmark have a unique and permanent civil registration number that enables individual linkage to nationwide administrative registers.
      • Schmidt M.
      • Pedersen L.
      • Sørensen H.T.
      The Danish civil registration system as a tool in epidemiology.
      Patients were identified from procedure codes in the Danish Vascular Registry. These data were merged with data from the Danish Anaesthesia Database, the Danish National Prescription Database and the National Patient Registry.

      The danish vascular registry

      This study is based on data from this nationwide database, founded in 1989, now comprising more than 190 000 procedures. Annual reports, evaluating all Danish vascular departments by a set of quality indicators, are accessible online (Karbase.dk).

      Anaesthesia type

      GA was defined as anaesthesia involving controlled mechanical ventilation. GA in combination with adjunctive RA was defined as GA. RA included solely spinal, epidural and peripheral nerve blocks without airway management. The type of anaesthesia (GA or RA), and the dates of surgery and hospital discharge registered in the Danish Vascular Registry were crosschecked with registrations in the Danish Anaesthesia Database and the National Patient Registry.

      Comorbidity

      Comorbidity, registered in the Danish Vascular Registry, includes diabetes, previous stroke or transitory cerebral ischaemia, hypertension, cardiac disease (previous myocardial infarction, ischaemic heart disease, valvular disease, and dysrhythmias), pulmonary disease, chronic dialysis, severe liver disease, dementia, and cancer.
      The Charlson Comorbidity Index (CCI) was generated from data in the National Patient Registry and includes age, diabetes, liver disease, malignancy, acquired immune deficiency syndrome, moderate to severe chronic kidney disease, chronic heart failure, chronic obstructive pulmonary disease, cerebrovascular accident, dementia, hemiplegia, connective tissue disease, and peptic ulcer disease.
      • Christensen S.
      • Johansen M.B.
      • Christiansen C.F.
      • Jensen R.
      • Lemeshow S.
      Comparison of Charlson comorbidity index with SAPS and Apache scores for prediction of mortality following intensive care.
      The definition of CCI, applied in this study, is reported in the online supplementary material (Supplementary file 1).
      The American Society of Anesthesiologists (ASA) score describes patients’ pre-operative status and is registered in the Danish Vascular Registry and the Danish Anaesthesia Database by an anaesthetist during the pre-operative anaesthetic consult.

      Anticoagulant medications

      The Danish National Prescription Database provides information regarding all prescription drugs sold in Danish community pharmacies on an individual level, linked to the national civil registration number.
      • Kildemoes H.W.
      • Sørensen H.T.
      • Hallas J.
      The Danish national prescription resgistry.
      Information was retrieved on the use of new oral anticoagulants, vitamin K antagonists, antiplatelet therapy, and low molecular weight heparin (Supplementary file 2) up to 180 days before surgery.

      Surgical data

      The Danish Vascular Registry provides information on procedure code, indication for surgery, urgency of surgery, anaesthesia type, duration of surgery, and estimated blood loss.

      Post-operative complications

      The Danish Vascular Registry provides information on wound complications, surgical complications, and general complications, defined in Table 1.
      Table 1Description of complications provided by the Danish Vascular Registry
      Post-operative complicationsDefinition
      General medical complications
       CardiacMyocardial infarction, cardiac failure, dysrhythmias
       PulmonaryPulmonary complication requiring treatment
       Cerebrovascular eventTransitory cerebral ischaemia, cerebral infarction, or haemorrhage
       Acute renal failureNeed for dialysis
       Intensive care unitIntensive care unit length of stay >3 days
      Surgical complications
       Compartment syndromeFasciotomy required
       Peripheral nerve lesionCausing motor or sensory deficit
       Peripheral embolisationSymptomatic embolisation distal to the arteriotomy
       BleedingRequiring surgery
      Wound complications
       Wound infectionDeep wound infection, including infection of the prosthesis
       Wound necrosisRequiring revision
       Wound lymphocele/-rrhoea>2 days of lymphorrhoea
       Wound haematomaRequiring revision

      Mortality

      The Danish Vascular Registry provides information on in hospital and 30 day all cause mortality and time to death via a link to the Danish civil registration system, holding information on date of birth, death, sex, and migration.
      • Schmidt M.
      • Pedersen L.
      • Sørensen H.T.
      The Danish civil registration system as a tool in epidemiology.

      Length of stay

      The Danish National Patient Registry holds information on all hospitalisations since 1977, including admission and discharge dates, discharge diagnosis, procedure codes, and surgical procedures.
      • Lynge E.
      • Sandegaard J.L.
      • Rebolj M.
      The Danish national patient register.
      Length of stay was defined as the number of days from the date of surgery to the date of discharge from hospital to home or to a nursing facility. Transfers to other specialty wards, in or outside the primary hospital, were included in the length of stay.

      Statistics

      Anticipating a 4% mortality rate
      • Nowygrod R.
      • Egorova N.
      • Greco G.
      • Anderson P.
      • Gelijns A.
      • Moskowitz A.
      • et al.
      Trends, complications, and mortality in peripheral vascular surgery.
      after GA, a power calculation was performed, estimating a 25% decrease in the RA group to a mortality rate of 3%. With an α of 0.05 and 80% power (1-β), 5 301 procedures were needed in each group. Knowing that only approximately one third of procedures are performed using RA, it was chosen to include all procedures from 1 January 2005 to 31 December 2017.
      Categorical variables were presented as percentages and frequencies and continuous variables using medians and first to third quartiles (Q1–Q3). The association between type of anaesthesia and hospital length of stay was examined using linear regression. Hospital length of stay was log transformed, and the assumption of linearity was assessed using QQ plots. Similarly, but using logistic regression, associations between type of anaesthesia and 30 day mortality, surgical, general and wound complications were analysed, and odds ratios presented in Forest plots. Mortality by type of anaesthesia was presented using Kaplan–Meier estimates. Data management was performed with SAS version 9.4 (SAS Institute Inc.) and analyses using R statistical software package version 3.5.0 (R Development Core Team).
      R Core Team. R
      A language and environment for statistical computing.
      Propensity score matching was used to identify a set of matched GA/RA pairs so that confounding by indication, associated with the observational study, could be minimised. The propensity score was derived by developing a logistic regression model that predicted the probability that a given patient would receive GA on the basis of age, cardiac and pulmonary disease, previous stroke, cancer, treatment with anticoagulant medication, procedure type (bypass, TEA or other), and urgency of surgery. Stratified matching was performed in four groups based on ASA score I–II, ASA score III–V, female, and male sex. The propensity score was used to match patients on a one to one basis to minimise the overall distance in propensity scores between the groups. The GA and RA patients were matched unless their estimated log odds from the logistic regression model were more than two standard deviations apart.

      Results

      A total of 18 450 first time procedures was identified in the Danish Vascular Registry. Of these procedures, 1 091 were excluded because of one or more of the following: temporary replacement civil registration numbers (n = 56), missing data on anaesthesia method (n = 376) or registered local infiltration analgesia performed by the surgeon (n = 765), leaving 17 359 procedures for analyses. The most frequently performed procedures were bypass surgery from the femoral artery to the popliteal or crural arteries, followed by thrombo-endarterectomy surgery of the femoral artery and its branches (Supplementary file 3).
      A total of 6 267 pairs were suitable for propensity score matching.
      Unmatched and matched patient characteristics and peri-operative data are presented in Table 2. GA was used in 61% of patients. More octogenarians were represented in the RA group. Also, in the GA group, cardiac and cerebrovascular comorbidity were more prevalent. Previous myocardial infarction was seen in 11.2% of patients in the GA group vs. 8.6% in the RA group (p < .001) and previous stroke in 8.5% of GA patients vs. 6.9% of RA patients (p < .001). Equivalently, more patients were treated with anticoagulant medication in the GA group (Table 2). More patients were ASA group 3–4 in this group. The difference in CCI was small but statistically significant (Table 2). No difference was found in the prevalence of diabetes, which was present in 24.8% of all patients.
      Table 2Patient characteristics and peri-operative data in the unmatched and matched cohort of patients in the Danish Vascular Registry operated on under general anaesthesia (GA) or regional anaesthesia (RA) for open inguinal and infra-inguinal arterial reconstructions in 2005–2017
      Unmatched GA (n = 10 509)Unmatched RA (n = 6 850)p valueMatched GA (n = 6 267)Matched RA (n = 6 267)p value
      Age – y70.0 (63.0, 77.0)72.0 (65.0, 78.0)<.00171.0 (64.0, 78.0)71.0 (64.0, 78.0).32
      Age > 80 y1 718 (16.3)1 265 (18.5)<.0011 076 (17.2)1 082 (17.3).91
      Female4 199 (39.9)2 720 (39.7).762 505 (40.0)2 505 (40.0)1.0
      ASA physiology score III–V4 813 (45.8)2 511 (36.7)<.0012 361 (37.7)2361 (37.7)1.0
      Charlson Comorbidity Score
       Low185 (1.8)29 (0.4)67 (1.1)29 (0.5)
       Moderate4 677 (44.5)3 249 (47.4)3 005 (47.9)3 008 (48.0)
       Severe5 647 (53.7)3 572 (52.1)<.0013 195 (51.0)3 230 (51.5)<.001
      Comorbidity
       Cerebrovascular894 (8.5)473 (6.9)<.001440 (7.0)448 (7.2).81
       Cardiac4 266 (40.6)2 265 (33.1)<.0012 123 (33.9)2 174 (34.7).35
      Pulmonary comorbidity
       COPD1 493 (14.2)973 (14.2)873 (13.9)879 (14.0)
       History of dyspnoea211 (2.0)215 (3.1)<.001144 (2.3)150 (2.4).98
       Anticoagulant medication2 838 (27.0)970 (14.2)<.001918 (14.6)952 (15.2).41
      Surgery
       Emergency
      Performed within h.
      1 213 (11.5)203 (3.0)209 (3.3)198 (3.2)
       Urgent
      Performed within days.
      2 286 (21.8)1 153 (16.8)1 096 (17.5)1 106 (17.6)
       Elective7 007 (66.7)5 490 (80.1)<.0014 960 (79.1)4 962 (79.2).68
       Blood loss – mL250 (100, 450)200 (100, 400)<.001225 (100, 410)200 (100, 400)<.001
       Duration of surgery – min129 (90, 180)125 (92, 175).31131 (95, 180)125 (93, 180)<.001
      Indication for surgery
       Acute ischaemia1 625 (15.5)430 (6.3)518 (8.3)414 (7.5)
       Claudication2 427 (23.1)2 251 (32.9)1 716 (27.4)2 077 (33.2)
       Rest pain1 957 (18.7)1 339 (19.6)1 300 (20.8)1 202 (19.2)
       Ulceration or gangrene3 135 (29.9)2 347 (34.3)2 082 (33.3)2 122 (33.9)
       Other indication1 342 (12.8)466 (10.4)<.001642 (10.3)436 (7.0)<.001
      Procedure type
       TEA2 872 (27.3)1 991 (29.1)1 832 (29.2)1 838 (29.3)
       Bypass4 262 (40.6)3 160 (46.1)2 803 (44.7)2 814 (44.9)
       Other3 375 (32.1)1 699 (24.8)<.0011 632 (26.0)1 615 (25.8).96
      Data are presented as n (%) or median (interquartile range). Numbers vary slightly depending on the quantity of missing information, range 0–2.5%. GA = general anaesthesia; RA = regional anaesthesia; ASA = American Society of Anesthesiologists; COPD = chronic obstructive pulmonary disease; TEA, thrombo-endarterectomy.
      Performed within h.
      Performed within days.
      Indication for surgery varied between the groups and more patients in the GA group needed emergency surgery. Accordingly, acute ischaemia was more prevalent in the GA group (Table 2).
      Univariable comparison of post-operative complications, using unmatched and matched data are presented in Table 3. Multivariable regression analysis of post-operative outcome, using unmatched and matched data are presented in Table 4 and Figure 1, Figure 2.
      Table 3Univariable analyses of complications in the unmatched and matched cohort of patients in the Danish Vascular Registry operated under general anaesthesia (GA) or regional anaesthesia (RA) for open inguinal and infra-inguinal arterial reconstructions in 2005–2017
      Post-operative complicationsUnmatched GA n = 10 509Unmatched RA n = 6 850p valueMatched GA n = 6 267Matched RA n = 6 267p value
      Cardiac295 (2.9)123 (1.8)<.001175 (2.8)113 (1.8).002
      Pulmonary172 (1.7)61 (0.9)<.001101 (1.6)55 (0.9).002
      Cerebrovascular event51 (0.5)17 (0.3).02821 (0.4)15 (0.3).69
      Acute renal failure, need for dialysis37 (0.4)11 (0.2).02518 (0.3)10 (0.2).18
      ICU >3 days70 (0.7)11 (0.2)<.00133 (0.5)11 (0.2).001
      Compartment syndrome47 (0.5)10 (0.1)<.00115 (0.2)10 (0.2).41
      Peripheral nerve lesion30 (0.3)19 (0.3).8519 (0.3)17 (0.3).84
      Peripheral embolisation30 (0.3)13 (0.2).2618 (0.3)13 (0.2).45
      Bleeding, requiring surgery205 (2.0)94 (1.4).003128 (2.1)86 (1.4).003
      Wound infection200 (1.9)116 (1.7).34119 (1.9)106 (1.7).50
      Wound necrosis195 (1.9)128 (1.9).97108 (1.8)115 (1.9).75
      Wound lymphocele/-lymphorrhoea954 (9.3)626 (9.2).83629 (10.3)563 (9.1).026
      Wound haematoma344 (3.4)214 (3.2).48205 (3.3)200 (3.2).75
      Data are presented as n (%). The description of complications can be found in Table 1. Numbers vary slightly depending on the quantity of missing information, range 0–3.1%. GA = general anaesthesia; RA = regional anaesthesia; ICU = intensive care unit.
      Table 4Outcome data in the unmatched and matched cohort of patients in the Danish Vascular Registry operated on under general anaesthesia (GA) or regional anaesthesia (RA) for open inguinal and infra-inguinal arterial reconstructions from 2005 to 2017
      OutcomeUnmatched GA (n = 10 509)Unmatched RA (n = 6 850)p valueMatched GA (n = 6 267)Matched RA (n = 6 267)p value
      30 day mortality429 (4.1)166 (2.4)<.001194 (3.1)150 (2.4).019
      1 y amputation rate1 033 (10.0)550 (8.2)<.001546 (8.8)515 (8.4).39
      Wound complications1 533 (15.0)969 (14.3).20955 (15.6)879 (14.2).030
      General complications723 (7.0)285 (4.2)<.001400 (6.5)261 (4.2)<.001
      Surgical complications408 (4.0)174 (2.6)<.001232 (3.8)157 (2.5)<.001
      Length of stay, days5 (3, 11)5 (3, 9)<.0015 (3, 9)5 (3, 8).14
      Data are presented as n (%) or median (interquartile range). The description of complications can be found in Table 1. Numbers vary slightly depending on the quantity of missing information, range 0–3.1%.
      Figure 1
      Figure 1Forest plot demonstrating odds ratio (95% confidence interval) for 30 day mortality, complications, one year amputation rate, and length of stay, after regional anaesthesia (RA) vs. general anaesthesia (GA) for open infra-inguinal arterial reconstruction surgery in the unmatched cohort, comprising 17 359 procedures in the Danish Vascular Registry.
      Figure 2
      Figure 2Forest plot demonstrating odds ratio (95% confidence interval) for 30 day mortality, complications, one year amputation rate, and length of stay, after regional anaesthesia (RA) vs. general anaesthesia (GA) for open infra-inguinal arterial reconstruction surgery in the propensity score matched cohort, comprising 6 267 + 6 267 procedures in the Danish Vascular Registry.
      In unmatched analyses, 30 day mortality and the incidence of general and surgical complications were significantly higher in the GA group. All types of general medical complications were seen more frequently in the GA group. Of the surgical complications, only post-operative bleeding was significantly more frequent in the GA group (Table 3).
      After matching, all categories of complications were more frequent, and the 30 day mortality rate was significantly higher in the GA group after multiple regression analysis (Table 4 and Fig. 2).
      The overall 30 day mortality was 3.4%, and incidence of general and surgical complications was 5.9% and 3.4%, respectively. Median length of stay was five (3, 10) days. The overall same side one year amputation rate was 9.3%.
      Kaplan–Meier curves, demonstrating matched and unmatched 30 day mortality, are presented in Fig. 3. Missing data are reported in the relevant tables. Numbers were small and this study focused on cases with complete data.
      Figure 3
      Figure 3Cumulative Kaplan–Meier estimate of mortality after open infra-inguinal arterial reconstruction surgery under general (GA) or regional anaesthesia (RA) in (A) unmatched and (B) propensity score matched cohorts in the Danish Vascular Registry.

      Discussion

      In this nationwide cohort of 17 359 open inguinal and infra-inguinal arterial revascularisation procedures, a significantly increased 30 day mortality and complication rate was demonstrated if the surgery was performed using GA compared with RA, in both unmatched and propensity score matched analysis.
      In 2016, Kehlet et al.
      • Kehlet M.
      • Jensen L.P.
      • Schroeder T.V.
      Risk factors for complications after peripheral vascular surgery in 3,202 patient procedures.
      evaluated a subpopulation of the present study cohort, comprising 3 202 patients undergoing open arterial revascularisation of the lower limb. One third of patients developed one or more complications. The 30 day mortality rate was 5% and one year mortality rate 15%. After one year, 19% were either amputated or dead. Of independent risk factors, they identified age, female gender, comorbidity, urgent surgery, and GA; anaesthesia type was the only modifiable factor. The present study included 13 years of data from the nationwide cohort and confirms that GA seems to affect outcome negatively.
      In a recent retrospective study from the Vascular Quality Initiative, Sgroi et al. presented data from 15 997 patients undergoing infra-inguinal bypass surgery from 2011 to 2016.
      • Sgroi M.D.
      • McFarland G.
      • Mell M.W.
      Utilization of regional versus general anesthesia and its impact on lower extremity bypass outcomes.
      Interestingly, only 3.5% of patients received RA, and this ratio decreased during the study period from 4.6% in 2011 to 2.6% in 2016, with a centre variability ranging from 0% to 30%. Despite the small number of patients in the RA group, they demonstrated a reduced length of stay and a lower frequency of both acute congestive heart failure and acute kidney injury and a trend towards lower mortality (1.1 vs. 2.2%, p = .07).
      From a clinical perspective, the choice of anaesthesia depends on various factors. In acute surgery, both anaesthetist and surgeon may prefer GA because of presumed difficulty of the procedure, long duration of surgery, and to avoid the potentially more time consuming procedure with RA prior to surgery. The anaesthesia personnel may lack regular experience in vascular surgery and, therefore, prefer GA. In addition, the risk of inadequate peripheral nerve blocks may also have implication for the decision.
      • Hu P.
      • Harmon D.
      • Frizelle H.
      Patient comfort during regional anesthesia.
      Thus, acute surgery will often be performed under GA. The present study used multiple regression and propensity score matching to adjust for these differences and confirmed the increased 30 day mortality and complication rate in the GA group.
      Ghanami et al. studied a cohort of 5 462 patients for elective infra-inguinal bypass procedures and found no difference in outcome after RA vs. GA.
      • Ghanami R.J.
      • Hurie J.
      • Andrews J.S.
      • Harrington R.N.
      • Corriere M.A.
      • Goodney P.P.
      • et al.
      Anesthesia-based evaluation of outcomes of lower-extremity vascular bypass procedures.
      Only 13% of procedures were under RA, but the complication rate was as high as 37%. They found increased trainee surgical involvement, transfusion need, and operating time in the GA group, perhaps indicating more complex surgical procedures. However, a fast procedure may also be associated with poor outcomes because of comorbidities and disease severity, that has no revascularisation potential. Also, procedures like femorocrural bypass have a long operating time, but the blood loss is usually low. Procedure type and urgency of surgery were included in the present statistical model to adjust for these variations.
      Cardiac complications may be prevented by avoiding the haemodynamic instability that can be induced by GA.
      • Gold M.S.
      • DeCrosta D.
      • Rizzuto C.
      • Ben-Harari R.R.
      • Ramanathan S.
      The effect of lumbar epidural and general anesthesia on plasma catecholamines and hemodynamics during abdominal aortic aneurysm repair.
      In the Danish Vascular Registry, cardiac complications are defined as acute myocardial infarction, congestive heart failure, or dysrhythmias. Cardiac complications were seen in 2.4% of patients in the present study and were significantly more common in the GA group in both unmatched and matched analyses.
      The incidence of a neuraxial haematoma after epidural or spinal blockade ranges from 1:168 000 to 1:190 000.
      • Moen V.
      • Dahlgren N.
      • Irestedt L.
      Severe neurological complications after centralneuraxial blockades in Sweden 1990–1999.
      More patients in the GA group received anticoagulant medication, platelet inhibitor, or low weight molecular heparin, which increases this risk significantly. In the present study, use of these medications was included in the propensity score matching model. Unfortunately, there is no available information about the indication for therapy, or whether the medication was discontinued before surgery.
      Avoiding mechanical ventilation may contribute to better outcome. Hausman et al. demonstrated that the complication rate, risk of pulmonary infection, prolonged length of mechanical ventilation, and unplanned post-operative intubation were reduced when using RA compared with GA in 2 644 patients with severe chronic obstructive pulmonary disease (COPD) undergoing a variety of surgical procedures.
      • Hausman M.S.
      • Jewell E.S.
      • Engoren M.
      Regional versus general anesthesia in surgical patients with chronic obstructive pulmonary disease.
      In the present study, in the unmatched cohort, a higher prevalence of patients with dyspnoea and patients with COPD was found in the RA group, suggesting that RA was preferred in this patient category. As expected, significantly more post-operative pulmonary complications were found after GA in the unmatched cohort, although the overall pulmonary complication rate was only 1.4%.
      Peripheral nerve blocks may be more favourable than neuraxial anaesthesia. A post hoc analysis from a large, randomised trial demonstrated that epidural analgesia was associated with increased cardiovascular morbidity.
      • Leslie K.
      • Myles P.
      • Devereaux P.
      • Williamson E.
      • Rao-Melancini P.
      • Forbes A.
      • et al.
      Neuraxial block, death and serious cardiovascular morbidity in the POISE trial.
      Peripheral nerve blocks have the advantage that they can be used despite anticoagulation therapy and severe cardiac comorbidity, such as aortic stenosis. Yazigi et al. randomised 50 patients for peripheral vascular surgery to GA or peripheral nerve blocks (combined sciatic and femoral nerve block) and found reduced frequency of intra-operative myocardial ischaemia in the RA group.
      • Yazigi A.
      • Madi-Gebara S.
      • Haddad F.
      • Hayeck G.
      • Tabet G.
      Combined sciatic and femoral nerve blocks for infrainguinal arterial bypass surgery: a case series.
      Furthermore, peripheral nerve blocks may have the potential to reduce post-operative chronic pain.
      • Borghi B.
      • DʼAddabbo M.
      • White P.F.
      • Gallerani P.
      • Toccaceli L.
      • Raffaeli W.
      • et al.
      The use of prolonged peripheral neural blockade after lower extremity amputation.
      The risk of peripheral nerve damage and masking of severe ischaemia, however, must be acknowledged when using RA in peripheral artery disease. No difference was found in the incidence of peripheral nerve damage. More cases of compartment syndrome and amputation were seen in the GA group; however, this was not reflected in the propensity score matched analysis and may be more related to the severity of limb ischaemia than to the type of anaesthesia.
      It would have been informative to investigate the development and frequency of peripheral nerve blocks during the 13 year study period. Anticoagulant medication is increasingly used, resulting in fewer procedures with neuraxial blockade. Peripheral nerve block procedures, however, have been an increasing part of standard anaesthesia practice in many hospitals during this period. Unfortunately, the distribution of peripheral and neuraxial nerve blocks is not provided by the Danish Vascular Registry. From the present study, it can be concluded only that avoiding GA may be beneficial.
      One strength of this study was the nationwide character of the Danish Vascular Registry, including 17 359 procedures. Validating data using the Danish Anaesthesia Database and the National Patient Registry further strengthens the results in terms of data purity. Propensity score matching with a matched sample size of 72% is acceptable and the risk of potential influence of incomplete matching can be considered small.
      • Yao X.I.
      • Wang X.
      • Speicher P.J.
      • Hwang E.S.
      • Cheng P.
      • Harpole D.H.
      • et al.
      Reporting and guidelines in propensity score analysis: a systematic review of cancer and cancer surgical studies.
      The validity of the Danish Vascular Registry has been evaluated repeatedly and more than 95% internal and external validity has been demonstrated.
      • Altreuter M.
      • Menyhei G.
      International validation of the Danish vascular registry Karbase: a Vascunet report.
      • Lausten J.L.
      • Jensen L.P.
      • Hansen A.K.
      Danish Vascular Registry
      Accuracy of clinical data in a population based vascular registry.
      • Lasota A.N.
      • Overvad K.
      • Eriksen H.H.
      • Tjønneland A.
      • Schmidt E.B.
      • Grønholdt M.L.M.
      Validity of peripheral arterial disease diagnoses in the Danish National Patient Registry.
      Data are, however, confined to the Danish population and healthcare system, which may impair generalisability.
      • Lees T.
      • Troeng T.
      • Thomson I.A.
      • Menyhei G.
      • Simo G.
      • Beiles B.
      • et al.
      International variations in infrainguinal bypass surgery - a VASCUNET report.
      Limitations of this study include that this is a retrospective cohort study. Confounding with selection and information bias may represent a potential problem. All data were registered by the vascular surgeon in charge of the operation, including registration of surgical complications, and some degree of under reporting may be expected. It was not possible to identify patients with unsuccessful or complicated procedures. Unfortunately, this information can be extrapolated only from registered complications and length of hospital stay, an endpoint that is somewhat insecure and debatable.
      • Boney O.
      • Moonesinghe S.R.
      • Myles P.S.
      • Grocott M.P.W.
      Standardizing endpoints in perioperative research.
      Additionally, the Danish Vascular Database does not provide detailed information about the timing of complications. It would be of interest to distinguish between intra-operative and post-operative complications, especially in terms of cardiac incidents, which were more common in the GA group. Whether this reflects intra-operative hypotension and hypoperfusion or insufficient post-operative pain control remains unanswered in the present study. Post-operative pain is affected by adding RA to GA. The frequency of combined GA and RA was registered by the surgeon, who may not have been aware of the importance of this specific information. As a result of poor validation and risk of severe registration bias, these data were not reported in the present study.
      Residual confounding should also be considered. It is difficult to compare patients affected by different pathologies (i.e. diabetic arteriopathy, aneurysmal disease, acute limb ischaemia), which may present with different severity of the disease, despite treatment probably being surgical in most cases. Neither the anaesthetist's nor the surgeon's reason for choosing RA or GA are available in these data. Also, the seniority of both anaesthetist and surgeon may have added important information to the study.
      The study period was 13 years, which may represent an additional limitation. During this period, endovascular treatment options have improved
      • Wrede A.
      • Wiberg F.
      • Acosta S.
      Increasing the elective endovascular to open repair ratio of popliteal artery aneurysm.
      and thus the most fragile patients may no longer need open surgery.
      The perspective of this hypothesis generating study is to bring attention to the importance of considering RA in patients scheduled for inguinal and infra-inguinal vascular surgery. Regular use of peripheral nerve block procedures and sedation may make RA more attractive. Large scale randomised clinical trials, evaluating the effect of GA vs. RA on outcome are needed, and can, according to these results, be considered safe and ethically justifiable.
      Neither the 2018 ESC/ESVS Guidelines on peripheral artery disease,
      • Aboyans V.
      • Ricco J.B.
      • Bartelink M.L.E.
      • Björck M.
      • Brodmann M.
      • Cohnert T.
      • et al.
      Editor's Choice - 2017 ESC guidelines on the diagnosis and treatment of peripheral arterial diseases, in collaboration with the European Society for Vascular Surgery (ESVS).
      nor the 2019 Global Vascular Guidelines on the Management of Chronic Limb Threatening Ischaemia, discuss the topic of GA vs. RA.
      • Conte M.S.
      • Bradbury A.W.
      • Kohl P.
      • White J.V.
      • Dick F.
      • Fitridge R.
      • et al.
      Global vascular guidelines on the management of chronic limb-threatening ischemia.
      According to the present results, this modifiable element of treating patients with chronic limb threatening ischaemia may deserve consideration in future guidelines and recommendations.
      In summary, in this nationwide cohort study, including 17 359 open inguinal and infra-inguinal arterial revascularisation procedures, GA was associated with a significantly higher 30 day mortality and more surgical and general complications than RA. The rate of wound complications, one year amputation rate, and length of stay were not affected.
      After propensity score matching of 6 267 pairs, 30 day mortality and rates of all types of complications were significantly higher in the GA group. Length of stay and one year amputation rate were not affected. In conclusion, RA may be associated with a better outcome, compared with GA, after open inguinal and infra-inguinal vascular surgery. In a clinical context when RA is not feasible, GA can still be considered as safe.

      Acknowledgements

      The authors wish to thank the Danish Vascular Registry, the Danish Anaesthesia Database, and the Danish Prescription Database for providing data for this project. Phillip Freeman, MD PhD, Department of Cardiology, Aalborg University Hospital, performed English language editing and proofreading.

      Appendix A. Supplementary data

      The following are the Supplementary data to this article:

      Conflict of interest

      None.

      Funding

      None.

      References

        • Nowygrod R.
        • Egorova N.
        • Greco G.
        • Anderson P.
        • Gelijns A.
        • Moskowitz A.
        • et al.
        Trends, complications, and mortality in peripheral vascular surgery.
        J Vasc Surg. 2006; 43: 205-216
        • Nehler M.R.
        • Duval S.
        • Diao L.
        • Annex B.H.
        • Hiatt W.R.
        • Rogers K.
        • et al.
        Epidemiology of peripheral arterial disease and critical limb ischemia in an insured national population.
        J Vasc Surg. 2014; 60: 686-695
        • Fleisher L.A.
        • Fleischmann K.E.
        • Auerbach A.D.
        • Barnason S.A.
        • Beckman J.A.
        • Bozkurt B.
        • et al.
        2014 ACC/AHA guideline on perioperative cardiovascular evaluation and management of patients undergoing noncardiac surgery: executive summary.
        Circulation. 2014; 130: 2215-2245
        • Neves S.E.
        Anesthesia for patients with peripheral vascular disease and cardiac dysfunction.
        Anesthesiol Clin. 2016; 34: 775-795
        • Johnson R.L.
        • Kopp S.L.
        • Burkle C.M.
        • Duncan C.M.
        • Jacob A.K.
        • Erwin P.J.
        • et al.
        Neuraxial vs general anaesthesia for total hip and total knee arthroplasty: a systematic review of comparative-effectiveness research.
        Br J Anaesth. 2016; 116: 163-176
        • Memtsoudis S.G.
        • Rasul R.
        • Suzuki S.
        • Poeran J.
        • Danninger T.
        • Wu C.
        • et al.
        Does the impact of the type of anesthesia on outcomes differ by patient age and comorbidity burden?.
        Reg Anesth Pain Med. 2014; 39: 112-119
        • Turek Z.
        • Sykora R.
        • Matejovic M.
        • Cerny V.
        Anesthesia and the microcirculation.
        Semin Cardiothorac Vasc Anesth. 2009; 13: 249-258
        • Barbosa F.T.
        • Jucá M.J.
        • Castro A.A.
        • Cavalcante J.C.
        Neuraxial anaesthesia for lower-limb revascularization.
        Cochrane Database Syst Rev. 2013; 7: CD007083
        • Sgroi M.D.
        • McFarland G.
        • Mell M.W.
        Utilization of regional versus general anesthesia and its impact on lower extremity bypass outcomes.
        J Vasc Surg. 2019; 69: 1874-1879
        • Gogarten W.
        • Vandermeulen E.
        • Van Aken H.
        • Kozek S.
        • Llau J.V.
        • Samama C.M.
        Regional anaesthesia and antithrombotic agents: recommendations of the European Society of Anaesthesiology.
        Eur J Anaesthesiol. 2010; 27: 999-1015
        • Yazigi A.
        • Madi-Gebara S.
        • Haddad F.
        • Hayeck G.
        • Tabet G.
        Combined sciatic and femoral nerve blocks for infrainguinal arterial bypass surgery: a case series.
        J Cardiothorac Vasc Anesth. 2005; 19: 220-221
        • Guimarães J. de F.
        • Angonese C.F.
        • Gomes R.K.
        • Junior V.M.
        • Farias C.
        Anesthesia for lower extremity vascular bypass with peripheral nerve block.
        Rev Bras Anesth. 2017; 67: 626-631
        • Selvi O.
        • Bayserke O.
        • Tulgar S.
        Use of femoral and sciatic nerve block combination in severe emphysematous lung disease for femoral popliteal arterial bypass surgery.
        Cureus. 2018; 10: 2-5
        • Kehlet M.
        • Jensen L.P.
        • Schroeder T.V.
        Risk factors for complications after peripheral vascular surgery in 3,202 patient procedures.
        Ann Vasc Surg. 2016; 36: 13-21
        • Singh N.
        • Sidawy A.N.
        • Dezee K.
        • Neville R.F.
        • Weiswasser J.
        • Arora S.
        • et al.
        The effects of the type of anesthesia on outcomes of lower extremity infrainguinal bypass.
        J Vasc Surg. 2006; 44: 964-970
        • Schmidt M.
        • Pedersen L.
        • Sørensen H.T.
        The Danish civil registration system as a tool in epidemiology.
        Eur J Epidemiol. 2014; 29: 541-549
        • Christensen S.
        • Johansen M.B.
        • Christiansen C.F.
        • Jensen R.
        • Lemeshow S.
        Comparison of Charlson comorbidity index with SAPS and Apache scores for prediction of mortality following intensive care.
        Clin Epidemiol. 2011; 3: 203-211
        • Kildemoes H.W.
        • Sørensen H.T.
        • Hallas J.
        The Danish national prescription resgistry.
        Scand J Public Health. 2011; 39: 38-41
        • Lynge E.
        • Sandegaard J.L.
        • Rebolj M.
        The Danish national patient register.
        Scand J Public Health. 2011; 39: 30-33
        • R Core Team. R
        A language and environment for statistical computing.
        R foundation for statistical computing. 2018 (Vienna, Austria)
        • Hu P.
        • Harmon D.
        • Frizelle H.
        Patient comfort during regional anesthesia.
        J Clin Anesth. 2007; 19: 67-74
        • Ghanami R.J.
        • Hurie J.
        • Andrews J.S.
        • Harrington R.N.
        • Corriere M.A.
        • Goodney P.P.
        • et al.
        Anesthesia-based evaluation of outcomes of lower-extremity vascular bypass procedures.
        Ann Vasc Surg. 2013; 27: 199-207
        • Gold M.S.
        • DeCrosta D.
        • Rizzuto C.
        • Ben-Harari R.R.
        • Ramanathan S.
        The effect of lumbar epidural and general anesthesia on plasma catecholamines and hemodynamics during abdominal aortic aneurysm repair.
        Anesth Analg. 1994; 78: 225-230
        • Moen V.
        • Dahlgren N.
        • Irestedt L.
        Severe neurological complications after centralneuraxial blockades in Sweden 1990–1999.
        Anesthesiology. 2004; 101: 950-959
        • Hausman M.S.
        • Jewell E.S.
        • Engoren M.
        Regional versus general anesthesia in surgical patients with chronic obstructive pulmonary disease.
        Anesth Analg. 2014; 120: 1405-1412
        • Leslie K.
        • Myles P.
        • Devereaux P.
        • Williamson E.
        • Rao-Melancini P.
        • Forbes A.
        • et al.
        Neuraxial block, death and serious cardiovascular morbidity in the POISE trial.
        Br J Anaesth. 2013; 111: 382-390
        • Borghi B.
        • DʼAddabbo M.
        • White P.F.
        • Gallerani P.
        • Toccaceli L.
        • Raffaeli W.
        • et al.
        The use of prolonged peripheral neural blockade after lower extremity amputation.
        Anesth Analg. 2010; 111: 1308-1315
        • Yao X.I.
        • Wang X.
        • Speicher P.J.
        • Hwang E.S.
        • Cheng P.
        • Harpole D.H.
        • et al.
        Reporting and guidelines in propensity score analysis: a systematic review of cancer and cancer surgical studies.
        J Natl Cancer Inst. 2017; 109: 1-9
        • Altreuter M.
        • Menyhei G.
        International validation of the Danish vascular registry Karbase: a Vascunet report.
        Eur J Vasc Endovasc Surg. 2019; 58: 609-613
        • Lausten J.L.
        • Jensen L.P.
        • Hansen A.K.
        • Danish Vascular Registry
        Accuracy of clinical data in a population based vascular registry.
        Eur J Vasc Endovasc Surg. 2004; 27: 216-219
        • Lasota A.N.
        • Overvad K.
        • Eriksen H.H.
        • Tjønneland A.
        • Schmidt E.B.
        • Grønholdt M.L.M.
        Validity of peripheral arterial disease diagnoses in the Danish National Patient Registry.
        Eur J Vasc Endovasc Surg. 2017; 53: 679-685
        • Lees T.
        • Troeng T.
        • Thomson I.A.
        • Menyhei G.
        • Simo G.
        • Beiles B.
        • et al.
        International variations in infrainguinal bypass surgery - a VASCUNET report.
        Eur J Vasc Endovasc Surg. 2012; 44: 185-192
        • Boney O.
        • Moonesinghe S.R.
        • Myles P.S.
        • Grocott M.P.W.
        Standardizing endpoints in perioperative research.
        Can J Anesth. 2016; 63: 159-168
        • Wrede A.
        • Wiberg F.
        • Acosta S.
        Increasing the elective endovascular to open repair ratio of popliteal artery aneurysm.
        Vasc Endovascular Surg. 2018; 52: 115-123
        • Aboyans V.
        • Ricco J.B.
        • Bartelink M.L.E.
        • Björck M.
        • Brodmann M.
        • Cohnert T.
        • et al.
        Editor's Choice - 2017 ESC guidelines on the diagnosis and treatment of peripheral arterial diseases, in collaboration with the European Society for Vascular Surgery (ESVS).
        Eur J Vasc Endovasc Surg. 2018; 55: 305-368
        • Conte M.S.
        • Bradbury A.W.
        • Kohl P.
        • White J.V.
        • Dick F.
        • Fitridge R.
        • et al.
        Global vascular guidelines on the management of chronic limb-threatening ischemia.
        Eur J Vasc Endovasc Surg. 2019; 58 (e33): S1-S109

      Linked Article

      • General Anaesthesia for Vascular Patients: An Endless Story of Love and Hate
        European Journal of Vascular and Endovascular SurgeryVol. 61Issue 3
        • Preview
          The continuous expansion of endovascular techniques has often made the percutaneous approach the only access sites needed during vascular procedures, even for complex cases. As a result, over the course of the last few decades, we have seen an ever decreasing number of vascular procedures performed under general anaesthesia on a global scale.
        • Full-Text
        • PDF
        Open Archive

      Comments

      Commenting Guidelines

      To submit a comment for a journal article, please use the space above and note the following:

      • We will review submitted comments as soon as possible, striving for within two business days.
      • This forum is intended for constructive dialogue. Comments that are commercial or promotional in nature, pertain to specific medical cases, are not relevant to the article for which they have been submitted, or are otherwise inappropriate will not be posted.
      • We require that commenters identify themselves with names and affiliations.
      • Comments must be in compliance with our Terms & Conditions.
      • Comments are not peer-reviewed.