Pre-operative predictors of poor outcomes in patients undergoing surgical lower extremity revascularisation – Retrospective cohort study

Accepted Manuscript Pre-operative predictors of poor outcomes in patients undergoing surgical lower extremity revascularisation – Retrospective cohort study Mohammed Ashrafi, Rohini Salvadi, Philip Foden, Stephanie Thomas, Mohamed Baguneid PII:

S1743-9191(17)30277-7

DOI:

10.1016/j.ijsu.2017.03.057

Reference:

IJSU 3692

To appear in:

International Journal of Surgery

Received Date: 6 February 2017 Revised Date:

21 March 2017

Accepted Date: 21 March 2017

Please cite this article as: Ashrafi M, Salvadi R, Foden P, Thomas S, Baguneid M, Pre-operative predictors of poor outcomes in patients undergoing surgical lower extremity revascularisation – Retrospective cohort study, International Journal of Surgery (2017), doi: 10.1016/j.ijsu.2017.03.057. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Pre-operative predictors of poor outcomes in patients undergoing surgical lower extremity revascularisation – retrospective cohort study a

b

c

RI PT

a

a*

Mohammed Ashrafi , Rohini Salvadi , Philip Foden , Stephanie Thomas , Mohamed Baguneid a

Department of Vascular and Endovascular Surgery, University Hospital of South Manchester,

SC

Manchester, UK b

Department of Medical Statistics, University Hospital of South Manchester, Manchester, UK

Department of Microbiology, University Hospital of South Manchester, Manchester, UK

M AN U

c

*Corresponding author at: Department of Vascular and Endovascular Surgery, University Hospital of South Manchester NHS Foundation Trust, Southmoor Road, Wythenshawe, Manchester, M23 9LT, UK. Tel.: +44 161 9987070. E-mail address: [email protected]

AC C

EP

TE D

Running title: Pre-operative predictors of poor outcomes in LER

1

ACCEPTED MANUSCRIPT

Abstract Background Surgical lower extremity revascularisation (LER) can lead to poor outcomes that

RI PT

include delayed hospital discharge, in-hospital mortality, major amputations and readmissions. The aim of this study was to identify pre-operative predictors

SC

associated with these poor clinical outcomes.

M AN U

Materials and methods

All patients (n=635; mean age 69; male 67.4%) who underwent surgical LER over a 5 year period in a single tertiary vascular institution were identified.

Patients

considered to have suffered a poor outcome (Group A) included all in-hospital

TE D

mortality and major amputations, delayed discharges with a length of stay (LOS) over one standard deviation above the mean or any readmission under any specialty within 12 months. Group A included 247 patients (38.9%) and the good outcome

EP

group included the remaining 388 patients (61.1%) from which a sample of 99

AC C

patients were selected as controls (Group B).

Results

Mean LOS for the entire study group was 14.4 ± 17.5 days, 12 month readmission rate was 29.1% and in-hospital mortality and major amputation rate was 2.7% and 1.4%, respectively. Pre-admission residence other than own home (OR 9.0; 95% CI 1.2-70.1; P=0.036), atherosclerotic disease burden (OR 2.2; 95% CI 1.3-3.8;

1

ACCEPTED MANUSCRIPT P=0.003) and tissue loss (OR 3.0; 95% CI 1.6-5.3; P<0.001) were identified as independent, statistically significant pre-operative predictors of poor outcome. Following discharge, group B patients had a significantly higher rate of amputation

RI PT

free survival and graft infection free survival (P<0.001) compared to group A.

Conclusion

SC

Recognition of pre-operative predictors of poor outcome should inform case

and post discharge follow up.

M AN U

selection and identify high risk patients requiring intensive perioperative optimisation

Keywords: surgical lower extremity revascularisation; poor outcomes; pre-operative

AC C

EP

TE D

predictors.

2

ACCEPTED MANUSCRIPT

1. Introduction Surgical lower extremity revascularisation (LER) with the use of a graft is considered a more durable form of revascularisation than endovascular treatment in patients

RI PT

with limb ischaemia [1]. However, it may be associated with significant perioperative and post–operative morbidity and mortality [2-3]. Common measures used to assess clinical outcome following surgical LER include length of stay (LOS), readmissions,

SC

perioperative mortality and perioperative amputation [4-8]. Furthermore, reduced LOS and decreased readmission rates are recognised markers of health care

M AN U

efficiency and quality [5]. Readmission rates following surgical LER can range from 12% to 23% [4, 9] and is associated with significant additional hospital costs [10]. Surgical LER is also associated with a perioperative mortality rate from 2 to 8% [11] and perioperative amputation rate of 2.6% [8]. These rates are even higher in

TE D

patients considered high risk from surgical intervention with mortality rates as high as 16% [12]. Various pre-operative factors have been shown to be associated with these poor clinical outcomes including chronic cardiac and respiratory disease,

EP

diabetes and renal impairment [4-5, 7-8]. Identification of these factors potentially allows for better case selection, perioperative optimisation and can permit more

AC C

accurate determination of clinical risk. However, no study to date has identified preoperative predictors of poor outcome using a composite of LOS, readmissions, perioperative mortality and perioperative amputation. Whilst, 30 day mortality and amputation rates have often been used to select the group of patients with a poor outcome [12]; they fail to acknowledge that prolonged LOS and readmissions are also poor outcome measures that behave as surrogate markers for quality [12].

3

ACCEPTED MANUSCRIPT The aim of this study was to identify, pre-operative predictors associated with poor clinical outcome in patients undergoing surgical LER with the use of a graft for all levels of ischaemia irrespective of cause. Poor clinical outcome was defined as the

AC C

EP

TE D

M AN U

SC

months, in-hospital mortality and major amputation.

RI PT

composite endpoint that includes delayed hospital discharge, readmissions within 12

4

ACCEPTED MANUSCRIPT

2. Material and Methods 2.1 Patient selection and data collection A retrospective cohort study at a single tertiary vascular and endovascular institution

RI PT

was undertaken. Data retrieval from electronic and case notes was performed on all patients’ undergoing surgical lower extremity revascularisation (LER) with the use of a graft over a 5 year period between January 2010 and December 2014. Patients

SC

were identified from the hospital episode statistics data and table 1 lists the Office of Population Censuses and Surveys (OPCS) codes used to identify patients based on

M AN U

their index procedure. Institutional board review was not required as this was a retrospective study of prospectively collected information on an electronic database regarding outcomes following procedures and reviewing patient case notes. There is no patient identifiable data presented and there is no way of identifying individual

TE D

patients and therefore patient consent was not required. The work has been reported in line with the STROBE criteria [13].

EP

2.2 Patient grouping

Patients were divided into two groups. Those considered to have a poor outcome

AC C

(group A) included all in-hospital mortality, in-hospital major amputations, delayed discharges with a length of stay (LOS) over one standard deviation above the mean or any hospital readmission under any speciality within 12 months. Major amputations included all amputations above the ankle joint. Patients who had none of the above were identified as the good outcome group. For comparison, a sample from the good outcome group was selected by ordering these cases in chronological order based on date of index admission and selecting cases to provide an even distribution over the 5 year time period to form the control group (group B). The 5

ACCEPTED MANUSCRIPT control group sample size was selected so that there was 80% power to detect differences of 15% between the groups (assuming 20-40% in the variables of interest for the poor outcome group) at the 5% significance level.

RI PT

2.3 Outcomes The primary outcome was defined as the identification of pre-operative factors associated with poor outcome. Secondary outcomes included post discharge

SC

mortality, major amputation free survival and graft infection free survival. Graft infection was diagnosed based on a combination of clinical, radiological and

M AN U

microbiological findings. 2.4 Follow up

The follow up period was calculated from the date of discharge following the index admission to the commencement of data collection for this study. Median follow up

TE D

for group A was 2.27 years (interquartile range (IQR); 1.29-3.78) and for group B was 2.03 years (IQR; 1.13-3.21).

EP

2.5 Data analysis

AC C

Continuous, normally distributed data were summarised as the mean ± standard deviation and categorical data were descried as counts and percentages. Univariate analyses were performed using chi-squared tests, Fisher’s exact tests and independent samples t-tests, as appropriate. Variables with a p-value of less than 0.2 were considered for inclusion in the initial multivariable logistic regression with poor outcome as the dependent variable. Eight variables were included in the multivariable logistic regression. The model with eight variables was analysed and the least statistically significant variable was removed and the model re-run. This

6

ACCEPTED MANUSCRIPT process was repeated until only variables that were statistically significant remained. A forward stepwise model selection procedure was used as a sensitivity analysis to the backward stepwise model selection procedure. A two-sided 5% significance level was used. Statistical analyses were performed using Statistical Package for Social

AC C

EP

TE D

M AN U

SC

RI PT

Sciences for Windows version 22.0 (SPSS, IBM, Armonk, NY, USA).

7

ACCEPTED MANUSCRIPT

3. Results 3.1 All patients A total of 635 patients (428 male and 207 female; mean age 69 ± 11.8 years)

RI PT

underwent 719 surgical LER procedures over the 5 year period. Mean LOS was 14.4 ± 17.5 (median: 9; IQR: 6-16) days and 185 (29.1%) patients were emergency admissions. Fifty nine (9.3%) patients had a LOS greater than one standard

SC

deviation above the mean. A total of 192 (30.3%) patients were readmitted under any speciality within 12 months of discharge from index admission. Of these 152

(2.7%) and 9 (1.4%), respectively. 3.2 Group A vs. Group B

M AN U

(79.2%) were due to a vascular problem. In-hospital mortality and amputation was 17

TE D

Group A consisted of 247 patients (38.9%) including all those considered to have had a poor outcome. The good outcome group included the remaining 388 patients (61.1%). Group B consisted of 99 patients which were randomly selected to provide

EP

sufficient power to detect differences of 15% in variables of interest between the poor outcome group and the good outcome group.

AC C

3.3 Primary outcome

3.3.1 Univariate analysis Pre-operative predictors of poor outcome were age (70.7 vs. 67.4 years; P=0.035), pre-admission residence other than own home (7.8% vs. 1%; P=0.016), previous lower limb vascular procedures (42.7% vs. 28.3%; P =0.013), atherosclerotic disease burden (defined as previously symptomatic LER, ischaemic heart disease, cerebrovascular and carotid disease, aneurysmal disease or mesenteric artery 8

ACCEPTED MANUSCRIPT disease) (46.4 vs. 28.3%; P=0.002), tissue loss (41.5% vs. 18.2%; P<0.001), critical limb ischaemia (intermittent claudication - 34.5% vs. 51.5%; critical limb ischaemia 51.5% vs. 28.3%; Other - 14.0% vs. 20.2%; P=0.001); emergency presentation (47.2% vs. 26.3%; P<0.001) and American Society of Anaesthesiologist’s (ASA)

RI PT

grade 4 (1 - 0.9% vs. 1.1%, 2 - 25.7% vs. 37.2%, 3 - 66.4% vs. 57.4%, 4 - 7.1% vs. 4.3%; P=0.037) (table 2).

SC

3.3.2 Multivariable analysis

The eight pre-operative factors included in the multivariable analysis were age, ASA,

M AN U

diabetes, tissue loss, emergency presentation, atherosclerotic disease burden, preadmission residence other than own home and indication. Pre-admission residence other than own home (OR 9.0; 95% CI 1.2-70.1; P=0.036), atherosclerotic disease burden (OR 2.2; 95% CI 1.3-3.8; P=0.003) and tissue loss (OR 3.0; 95% CI 1.6-5.3;

TE D

P<0.001) were independent statistically significant pre-operative predictors of poor outcome from multivariable analysis (figure 1). The same three variables were selected using a forward stepwise selection procedure.

EP

3.4 Procedures and in-hospital complications

AC C

Table 3 outlines details regarding the surgical LER procedures performed and the inhospital complications.

3.5 Secondary outcome Following discharge, group B had a significantly higher rate of major amputation free survival (99% vs. 83%; P<0.001) and graft infection free survival (100% vs. 85.7%; P<0.001) compared to group A at a median follow up time of 2.02 and 2.27 years, respectively (table 4).

9

ACCEPTED MANUSCRIPT

4. Discussion We defined poor clinical outcome post-surgical LER as a composite product of delayed discharge, readmission within 1 year, in-hospital mortality or in-hospital

RI PT

major amputation. The median LOS for all surgical LER procedures was 9 days in our institution. This is similar to McPhee et al who found in a similar group of patients a LOS between 10 and 12 days [4]. Ambler et al found a higher LOS of 11 days;

SC

however they only included patients with critical limb ischaemia (CLI) [14]. Both Damrauer et al and Ghanami et al reported shorter LOS of 4 and 6 days

M AN U

respectively, although their cohort of patients only included those undergoing infrainguinal LER procedures [5, 15]. We report a 12 month readmission rate of 30.3% in this cohort of patients. Previous studies have only assessed readmission up to 30 days and found the rate of readmission varied between 16 and 24.4% which we

TE D

would have expected to rise if they had been followed up for a longer period [4-5, 10, 16,-17]. Our In-hospital mortality rate of 2.7% fell within the range of previously reported studies in similar cohorts which range from 1.6-3.1% [4, 18-20]. In-hospital

EP

amputation rate was less than half that reported by Sachs et al; however their series was significantly larger with over 250,000 patients undergoing surgical LER

AC C

procedures over a 9 year period [21].

We identified, using univariate analysis, 8 independent pre-operative factors associated with poor clinical outcome of which multivariable analysis identified preadmission residence other than own home, atherosclerotic disease burden and tissue loss as independent significant pre-operative predictors of poor outcome (figure 2). Age has previously been identified as a predictor of poor outcome in 10

ACCEPTED MANUSCRIPT patients undergoing surgical LER [22]. Our findings concur with previous studies and it would therefore be prudent to ensure patient over 70 years undergo intensive preoperative optimisation, as they are more likely to have a poor outcome post-surgical LER [7, 23-24]. With centralisation of vascular services in the United Kingdom, the

RI PT

number of admissions to vascular institutions from other non-vascular specialist hospitals is increasing. Our results indicate that pre-admission residence from other than the patient’s own home is a detrimental factor to good clinical outcome. This is

SC

in keeping with earlier findings by Feinglass et al who identified nursing home residence pre-admission conferred greater risk of poor outcome [8]. As this is

M AN U

becoming more common, strategies need to be devised to optimise this sub-group of patients. As expected, emergency presentations, those who have had previous endovascular or surgical LER, those with atherosclerotic disease burden and higher ASA grades had poorer post-operative outcomes [5, 8, 10, 16, 17, 23, 25-26].

TE D

Identifying this group of patients earlier and undertaking frequent assessment would allow optimisation of their co-morbidities as well as reducing the likelihood of these patients requiring emergency surgery. Both critical limb ischaemia and tissue loss

EP

were independent pre-operative predictors of poor outcome in our series, again

AC C

highlighting the need to identify at risk patients earlier to avoid progression to these severe states. Although, tissue loss has previously been identified as a factor associated with readmission post-surgical LER [4, 16] and protracted LOS [27] we have identified it as a predictor of in-hospital mortality and in-hospital amputation as well. Our findings of critical limb ischaemia as a predictor of poor outcome agrees with Simon et al who found LOS was prolonged and in-hospital mortality was higher in patients with CLI [20].

11

ACCEPTED MANUSCRIPT BMI, gender, diabetes, smoking, hypertension, dyslipidaemia and renal impairment were not associated with poor outcome post-surgical LER, some of which are in contrast to previous findings. Siracuse et al and Zhang et al have shown obesity is associated with prolonged LOS and readmission, respectively [10, 27] and BMI has

RI PT

been shown to be linked to perioperative mortality [22]. Female gender has been shown to be associated with higher readmission rates [4] and in combination with race has been found to be associated with delayed discharge [28]. However, other

SC

studies have agreed with our findings that gender alone does not affect outcome [18, 28]. Diabetes has been linked previously to increased rate of readmissions post-

M AN U

surgical LER [5, 17], however, in agreement with our findings, Wallaert et al found no association between diabetes and in-hospital mortality [19]. McPhee et al found current smokers were more likely to be readmitted post-surgical LER, however, we

TE D

found no such association with smoking and poor outcome [29].

Perioperative blood transfusion has previously been shown to be predictive of

EP

morbidity post-surgical LER [30] and our results suggest administration of blood during the index admission is associated with poor clinical outcome. Although we

AC C

found no significant difference between the poor and good outcome groups with regards to wound complications, we identified that in-hospital wound infection is linked to a poor clinical outcome in keeping with previous findings which suggested wound infections were predictors of readmission [4, 16]. Davies et al found that discharge disposition to home predicts better outcome in patients undergoing endovascular LER [31]. Our findings agree with this as discharge location other than own home along with in-hospital vascular complication were found to be associated with poor clinical outcome. 12

ACCEPTED MANUSCRIPT

During the follow up period, of those patients discharged, there was a significantly higher rate of major amputation and graft infection in the poor outcome group. As a

RI PT

measure to reduce post-operative graft infection rates, we suggest more frequent and intensive follow up and a prolonged antibiotic prophylactic course in those patients identified to be at high risk of suffering a poor outcome based on preoperative and in-hospital factors outlined above. In order to reduce the high rate of

SC

major amputations in the poor outcome group, we recommend the need for intensive

M AN U

pre-operative optimisation and time efficient management of these patients.

The study had certain limitations. Firstly, it was retrospective in nature. Secondly, we utilised only a representative sample of the good outcome group. Although we were

TE D

confident this was an accurate representation of the good outcome group, there is a possibility that the data presented for this sub-group of patients does not correlate with the whole population of the good outcome group. Lastly, there were certain

EP

factors which have previously been reported to be associated with predicting poor

AC C

outcome in surgical LER patients which our data did not allow us to assess. These include ethnicity, pre- and post-operative mobility, independence and cognitive impairment [5-6, 14, 28, 32-34].

13

ACCEPTED MANUSCRIPT

5. Conclusions In conclusion, recognition of pre-operative predictors of poor outcome should inform case selection and identify high risk patients requiring intensive perioperative

RI PT

optimisation and post discharge management and follow up. We also recommend identification and intervention in patients at the stage of pre-critical limb ischaemia in order to reduce the incidence of patients presenting with critical limb ischaemia,

AC C

EP

TE D

M AN U

SC

tissue loss and as emergency cases, all of which are predictors of poor outcomes.

14

ACCEPTED MANUSCRIPT

Author disclosure None.

RI PT

Source of funding This work was supported by an unrestricted educational grant from Pfizer UK.

Conflicts of interest

AC C

EP

TE D

M AN U

SC

None.

15

ACCEPTED MANUSCRIPT

References 1.

DJ Adam, JD Beard, T Cleveland, J Bell, AW Bradbury, JF Forbes, FG Fowkes, I Gillepsie,

CV Ruckley, G Raab, H Storkey, Bt participants. Bypass versus angioplasty in severe ischaemia of the leg (BASIL): multicentre, randomised controlled trial. Lancet. 366 (9501) (2005) 1925-1934. GM LaMuraglia, MF Conrad, T Chung, M Hutter, MT Watkins, RP Cambria. Significant

RI PT

2.

perioperative morbidity accompanies contemporary infrainguinal bypass surgery: an NSQIP report. J Vasc Surg. 50 (2) (2009) 299-304. 3.

MS Conte, DF Bandyk, AW Clowes, GL Moneta, H Namini, L Seely. Risk factors, medical

multicenter trial. J Vasc Surg. 42(3) (2005)456-464.

JT McPhee, NR Barshes, KJ Ho, A Madenci, CK Ozaki, LL Nguyen, M Belkin. Predictive

M AN U

4.

SC

therapies and perioperative events in limb salvage surgery: observations from the PREVENT III

factors of 30-day unplanned readmission after lower extremity bypass. J Vasc Surg. 57 (4) (2013) 955-962. 5.

SM Damrauer, AC Gaffey, A DeBord Smith, RM Fairman, LL Nguyen. Comparison of risk

(5) (2015) 1192-1200. 6.

TE D

factors for length of stay and readmission following lower extremity bypass surgery. J Vasc Surg. 62

TC Collins, M Johnson, J Daley, WG Henderson, SF Khuri, HS Gordon. Preoperative risk

factors for 30-day mortality after elective surgery for vascular disease in Department of Veterans

7.

EP

Affairs hospitals: is race important? J Vasc Surg. 34 (4) (2001) 634-640. PP Goodney, BW Nolan, A Schanzer, J Eldrup-Jorgensen, AC Stanley, DH Stone, DS

AC C

Likosky, JL Cronenwett, VSGONN England. Factors associated with death 1 year after lower extremity bypass in Northern New England. J Vasc Surg. 51 (1) (2010) 71-78. 8.

J Feinglass, MW Sohn, H Rodriguez, GJ Martin, WH Pearce. Perioperative outcomes and

amputation-free survival after lower extremity bypass surgery in California hospitals, 1996-1999, with follow-up through 2004. J Vasc Surg. 50 (4) (2009) 776-783. 9.

TR Vogel, RL Kruse. Risk factors for readmission after lower extremity procedures for

peripheral artery disease. J Vasc Surg. 58 (1) (2013) 90-97. 10.

JQ Zhang, T Curran, JC McCallum, L Wang, MW Wyers, AD Hamdan, RJ Guzman, ML

Schermerhorn. Risk factors for readmission after lower extremity bypass in the American College of Surgeons National Surgery Quality Improvement Program. J Vasc Surg. 59 (5) (2014) 1331-1339. 16

ACCEPTED MANUSCRIPT 11.

L Norgren. [New international consensus document on peripheral arterial disease. TASC II for

improved care]. Lakartidningen. 104 (19) (2007) 1474-1475. 12.

E Saarinen, S Vuorisalo, P Kauhanen, A Albäck, M Venermo. The benefit of revascularization

in nonagenarians with lower limb ischemia is limited by high mortality. Eur J Vasc Endovasc Surg. 49 (4) (2015) 420-425. E Elm, D Altman, M Egger, S Pocock, P Gøtzsche, J Vandenbroucke. The strengthening the

RI PT

13.

reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies. Int. J. Surg. 12 (2014) 1495–1499.

GK Ambler, A Dapaah, N Al Zuhir, PD Hayes, MS Gohel, JR Boyle, K Varty, PA Coughlin.

SC

14.

Independence and mobility after infrainguinal lower limb bypass surgery for critical limb ischemia. J Vasc Surg. 59 (4) (2014) 983-987.

RJ Ghanami, J Hurie, JS Andrews, RN Harrington, MA Corriere, PP Goodney, KJ Hansen,

M AN U

15.

MS Edwards. Anesthesia-based evaluation of outcomes of lower-extremity vascular bypass procedures. Ann Vasc Surg. 27 (2) (2013) 199-207. 16.

F Aziz, EB Lehman, AB Reed. Unplanned return to operating room after lower extremity

687. 17.

TE D

arterial bypass is an independent predictor for hospital readmission. J Vasc Surg. 263 (3) (2016) 678-

A Najafian, S Selvarajah, ER Schneider, MB Malas, BA Ehlert, KC Orion, AH Haider, CJ

Abularrage. Thirty-day readmission after lower extremity bypass in diabetic patients. J Surg Res. 200

18.

EP

(1) (2016) 356-364.

RP Duffy, JE Adams, PW Callas, A Schanzer, PP Goodney, MA Ricci, JL Cronenwett, DJ

AC C

Bertges, VSGoN England. The influence of gender on functional outcomes of lower extremity bypass. J Vasc Surg. 60 (5) (2014) 1282-1290. 19.

JB Wallaert, BW Nolan, J Adams, AC Stanley, J Eldrup-Jorgensen, JL Cronenwett, PP

Goodney. The impact of diabetes on postoperative outcomes following lower-extremity bypass surgery. J Vasc Surg. 56 (5) (2012) 1317-1323. 20.

JP Simons, A Schanzer, BW Nolan, DH Stone, JA Kalish, JL Cronenwett, PP Goodney,

VSGoN England. Outcomes and practice patterns in patients undergoing lower extremity bypass. J Vasc Surg. 55 (6) (2012) 1629-1636.

17

ACCEPTED MANUSCRIPT 21.

T Sachs, F Pomposelli, A Hamdan, M Wyers, M Schermerhorn. Trends in the national

outcomes and costs for claudication and limb threatening ischemia: angioplasty vs bypass graft. J Vasc Surg. 54 (4) (2011) 1021-1031. 22.

KA Giles, AD Hamdan, FB Pomposelli, MC Wyers, JJ Siracuse, ML Schermerhorn. Body

mass index: surgical site infections and mortality after lower extremity bypass from the National

23.

RI PT

Surgical Quality Improvement Program 2005-2007. Ann Vasc Surg. 24 (1) (2010) 48-56.

VK Moitra, BC Flynn, M Mazzeffi, C Bodian, D Bronheim, JE Ellis. Indication for surgery, the

revised cardiac risk index, and 1-year mortality. Ann Vasc Surg. 25 (7) (2011) 902-908.

TC Collins, D Nelson, JS Ahluwalia. Mortality following operations for lower extremity

SC

24.

peripheral arterial disease. Vasc Health Risk Manag. 6 (2010) 287-296. 25.

L Whittaker, LD Wijesinghe, DC Berridge, DJ Scott. Do patients with critical limb ischaemia

M AN U

undergo multiple amputations after infrainguinal bypass surgery? Eur J Vasc Endovasc Surg. 21 (5) (2001) 427-431. 26.

BW Nolan, RR De Martino, DH Stone, A Schanzer, PP Goodney, DW Walsh, JL Cronenwett

JL, VSGoN England. Prior failed ipsilateral percutaneous endovascular intervention in patients with critical limb ischemia predicts poor outcome after lower extremity bypass. J Vasc Surg. 54 (3) (2011)

27.

TE D

730-735.

JJ Siracuse, HL Gill, DW Jones, DB Schneider, PH Connolly, I Parrack, ZS Huang, AJ

Meltzer. Risk factors for protracted postoperative length of stay after lower extremity bypass for critical

28.

EP

limb ischemia. Ann Vasc Surg. 28 (6) (2014) 1432-1438. AK Jain, CA Kalbaugh, MA Farber, WA Marston, R Vallabhaneni. Race and gender affect

29.

AC C

outcomes of lower extremity bypass. J Vasc Surg. 60 (5) (2014) 1275-1280. JT McPhee, LL Nguyen, KJ Ho, CK Ozaki, MS Conte, M Belkin. Risk prediction of 30-day

readmission after infrainguinal bypass for critical limb ischemia. J Vasc Surg. 57 (6) (2013) 14811488. 30.

SD O'Keeffe, DL Davenport, DJ Minion, EE Sorial, ED Endean, ES Xenos. Blood transfusion

is associated with increased morbidity and mortality after lower extremity revascularization. J Vasc Surg 51 (3) (2010) 616-621. 31.

MG Davies, HF El-Sayed. Objective performance goals after endovascular intervention for

critical limb ischemia. J Vasc Surg. 62 (6) (2015) 1555-1563.

18

ACCEPTED MANUSCRIPT 32.

N Ahmad, GN Thomas, C Chan, P Gill. Ethnic differences in lower limb revascularisation and

amputation rates. Implications for the aetiopathology of atherosclerosis? Atherosclerosis. 233 (2) (2014): 503-507. 33.

JS Partridge, M Fuller, D Harari, PR Taylor, FC Martin, JK Dhesi. Frailty and poor functional

18 (2015) 57-63. 34.

RI PT

status are common in arterial vascular surgical patients and affect postoperative outcomes. Int J Surg.

SM Taylor, JW York, DL Cull, CA Kalbaugh, AL Cass, EM Langan. Clinical success using

patient-oriented outcome measures after lower extremity bypass and endovascular intervention for

AC C

EP

TE D

M AN U

SC

ischemic tissue loss. J Vasc Surg. 50 (3) (2009) 534-541.

19

ACCEPTED MANUSCRIPT Table 1. OPCS codes used to identify patients from hospital episode statistics data who underwent surgical lower extremity revascularisation Procedure

L16

Extra-anatomic bypass of aorta

L20 (L20.4, L20.5, L20.6, L20.8, L20.9)

Other emergency bypass of segment of aorta

L21 (L21.4, L21.5, L21.6, L21.8, L21.9)

Other bypass of segment of aorta

L50

Other emergency bypass of iliac artery

L51

Other bypass of iliac artery

L52

Reconstruction of iliac artery

L53

Other open operations on iliac artery

L58

Other emergency bypass of femoral artery

L59

Other bypass of femoral artery

L60

Reconstruction of femoral artery

M AN U

SC

RI PT

OPCS code

Revision of reconstruction of artery

AC C

EP

TE D

L65

20

ACCEPTED MANUSCRIPT Table 2. Univariate analysis of pre-operative predictors of poor outcome Pre-operative factor

Group A

Group B

P value

Age (years)

70.7 +/- 10.8

67.4 +/- 14.2

0.035

Gender (Male)

150 (60.7%)

67 (67.7%)

0.23

25.9 +/- 5.4

26.4 +/- 4.6

0.49

18 (7.8%)

1 (1%)

0.016

2

BMI (kg/m ) Residence other than own home

0.11

No

158 (66.9%)

75 (76.5%)

Diet control

11 (4.7%)

7 (7.1%)

Tablet control

46 (19.5%)

12 (12.2%)

Insulin control

21 (8.9%)

4 (4.1%)

41 (16.6%)

15 (15.2%)

84 (34%)

33 (33.3%)

122 (49.4%)

51 (51.5%)

Hypertension

182 (77.4%)

71 (71.7%)

0.27

Dyslipidaemia

155 (70.5%)

73 (73.7%)

0.55

Alcohol intake > 28units/week

29 (12.6%)

14 (14.3%)

0.68

IHD

93 (39.6%)

34 (34.3%)

0.37

Atherosclerotic disease burden

103 (46.4%)

28 (28.3%)

0.002

Renal impairment (eGFR < 60)

64 (27.2%)

23 (23.2%)

0.45

71 (30.2%)

31 (31.3%)

0.84

Previous vascular intervention

100 (42.7%)

28 (28.3%)

0.013

Tissue loss

97 (41.5%)

18 (18.2%)

<0.001

116 (47.2%)

26 (26.3%)

<0.001

Smoker Ex-smoker

COPD/asthma

Indication

EP

Emergency admission

M AN U

Non smoker

SC

0.92

TE D

Smoking

RI PT

Diabetes

0.001

Intermittent claudication

81 (34.5%)

51 (51.5%)

Critical limb ischaemia

121 (51.5%)

28 (28.3%)

1 (0.4%)

2 (2%)

32 (13.6%)

18 (18.2%)

AC C

Trauma

Other (acute limb ischaemia,

iatrogenic, pseudoaneurysm, aneurysm, graft infection)

ASA

0.037 1

2 (0.9%)

1 (1.1%)

2

58 (25.7%)

35 (37.2%)

3

150 (66.4%)

54 (57.4%)

4

16 (7.1%)

4 (4.3%)

BMI – body mass index; IHD – ischaemic heart disease; COPD – chronic obstructive pulmonary disease; ASA - American Society of Anaesthesiologist’s.

21

ACCEPTED MANUSCRIPT

Table 3. Surgical lower extremity revascularisation procedures performed and inhospital complications Group B

P value

Antibiotic treatment during admission

124 (50.2%)

41 (41.4%)

0.14

Microbial prophylaxis intra-operatively

219 (88.7%)

91 (91.9%)

0.37 0.14

16 (6.8%)

14 (14.1%)

7 (3%)

3 (3%)

Common femoral artery reconstruction

67 (28.5%)

33 (33.3%)

femoral-femoral crossover

24 (10.2%)

14 (14.1%)

femoral-popliteal Above Knee

43 (18.3%)

femoral-popliteal Below Knee

37 (15.7%)

Femoral-crural

15 (6.4%)

SC

Index Procedure

RI PT

Group A

ileo-femoral

Axillo-femoral Other Bypass conduit prosthetic vein composite

3 (3%) 3 (3%)

13 (5.5%)

6 (6.1%)

184 (79%)

81 (82.7%)

48 (20.6%)

17 (17.3%)

1 (0.4%)

0

<1 hr

4 (1.8%)

0

1-2 hr

29 (12.7%)

13 (13.3%)

2-3 hr

53 (23.2%)

27 (27.6%)

>3 hr

142 (62.3%)

58 (59.2%)

37 (15.9%)

16 (16.2%)

EP

Re-do surgery

7 (7.1%)

13 (5.5%)

TE D

Duration of surgery

16 (16.2%)

M AN U

Aorto-femoral

Adjunctive procedures during admission

AC C

None

0.084

0.95 0.56

154 (66.1%)

70 (70.7%)

Angiogram

19 (8.2%)

4 (4%)

Angioplasty

29 (12.4%)

11 (11.1%)

Stent

31 (13.3%)

14 (14.1%)

84 (34%)

21 (21.2%)

0.019

86 (37.1%)

26 (26.3%)

0.057

Blood transfusion

Wound complication

0.028

Wound complication type Seroma

12 (14.1%)

10 (38.5%)

7 (8.2%)

0

Infection

50 (58.8%)

11 (42.3%)

Haematoma

16 (18.8%)

5 (19.2%)

32 (13.7%)

10 (10.1%)

Lymphatic

Chest complication

0.72

22

0.37

ACCEPTED MANUSCRIPT Cardiac complication

21 (8.9%)

4 (4%)

0.12

Renal complication

13 (5.5%)

3 (3%)

0.41

7 (3%)

0

0.11

Vascular complication

42 (17.9%)

2 (2%)

<0.001

Other medical complications

24 (10.2%)

6 (6.1%)

0.23

Other surgical complications

14 (6%)

2 (2%)

0.16

24 (10.8%)

0

0.003

Hospital acquired infection

AC C

EP

TE D

M AN U

SC

RI PT

Discharge location other than own home

23

ACCEPTED MANUSCRIPT Table 4. Secondary outcomes following discharge Group B

P value

5 (2.2%)

0 (0%)

0.329

mortality at end of follow up

41 (18.4%)

12 (12.1%)

0.16

Amputation free survival at 12 months

197 (87.9%)

99 (100%)

<0.001

Amputation free survival at end of follow up

186 (83.0%)

98 (99.0%)

<0.001

Graft infection free survival at end of follow up

192 (85.7%)

99 (100%)

<0.001

Amputation and/or graft infection free survival at end of follow up

163(72.8%)

98 (99.0%)

<0.001

AC C

EP

TE D

M AN U

SC

30 day mortality

RI PT

Group A

24

ACCEPTED MANUSCRIPT

Figure legend Figure 1. Multivariable logistic regression model for poor outcomes

RI PT

Figure 2. Summary of the significant findings of the study and subsequent

AC C

EP

TE D

M AN U

SC

preventative advice

25

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

Highlights •

Surgical lower extremity revascularisation can lead to poor post-operative patient outcomes. Pre-operative factors including residence, severity of ischaemia and

RI PT

•

atherosclerotic disease burden are independent predictors of poor outcome. The current results reinforce the need to recognise these pre-operative

SC

predictors of poor outcome to allow case selection and the identification of

EP

TE D

discharge follow up.

M AN U

high risk patients requiring intensive perioperative optimisation and post

AC C

•