- Email: [email protected]
Discharge on the First Postoperative Day after Elective Carotid Endarterectomy
Discharge on the First Postoperative Day after Elective Carotid Endarterectomy Pierre Van den Brande,1 Isolde Van Heymbeeck,1 Erik Debing,1 Dimitri Aerden,1 Karl von Kemp,1 Leslie Moerman,1 Chris Verborgh,2 and Patrick Haentjens,3 Brussels, Belgium
Background: Medical complications may prolong the hospital stay after elective carotid endarterectomy (CEA). We prospectively assessed the social and medical feasibility and safety of patient discharge on the first postoperative day after elective CEA and unplanned readmissions. Methods: Between June 2011 and January 2012, 57 consecutive patients scheduled for elective CEA were enrolled with the aim of discharge on the first postoperative day if there were no medical contraindications and on the condition that the patient should not be left alone during the first day and night at home. CEA was carried out under local or general anesthesia. After discharge, the patients were contacted to ascertain the occurrence of arterial hypertension, cerebral hyperperfusion, focal cerebral ischemia, or hospital readmission. Results: Sixty-two CEA were carried out in 57 patients (33 men and 24 women ranging in age from 51e89 years). The indications for CEA were: asymptomatic high grade stenosis in 27, hemispheric transient ischemic attack in 12, amaurosis fugax in 6, recovered stroke in 16, and nonlateralizing signs in 1. There were no cases of perioperative stroke or death. Discharge on the first postoperative day was achieved in 45 cases (73%). In 15 cases (24%), discharge was on the second postoperative day because of the absence of a relative (12 cases) or for medical reasons (3 cases). Discharge was on day 3 in 1 case, and on day 10 in another, both for medical reasons. No cases of severe arterial hypertension, stroke, mortality, or readmission for reasons related to the CEA procedure were recorded up to postoperative day 30. Conclusion: In this study, the majority of patients undergoing elective CEA were discharged safely on the first postoperative day. Social reasons, rather than medical reasons, underlied most cases of later discharge. There were no unplanned readmissions for complications of CEA.
INTRODUCTION Consistent with The North American Symptomatic Carotid Endarterectomy Trial,1 the Asymptomatic Carotid Artery Study,2 and the European Carotid 1 Department of Vascular Surgery, Universitair Ziekenhuis Brussel, Brussels, Belgium. 2 Department of Anesthesiology, Universitair Ziekenhuis Brussel, Brussels, Belgium. 3 Center for Outcome Research, Universitair Ziekenhuis Brussel, Brussels, Belgium.
Correspondence to: Pierre Van den Brande, MD, PhD, Universitair Ziekenhuis Brussel, Department of Vascular Surgery, Laarbeeklaan 101, B-1090 Brussels, Belgium; E-mail: [email protected] Ann Vasc Surg 2014; 28: 901–907 http://dx.doi.org/10.1016/j.avsg.2013.10.014 Ó 2014 Elsevier Inc. All rights reserved. Manuscript received: July 8, 2013; manuscript accepted: October 6, 2013.
Surgery Trial,3 carotid endarterectomy (CEA) has been proposed as a treatment option for carotid atherosclerotic disease in proportion to the degree of stenosis because of its beneficial role in the prevention of stroke. The replacement of selective catheter angiography for diagnostic assessment by noninvasive imaging approaches, such as magnetic resonance angiography, computed tomography angiography, and Doppler ultrasonography have resulted in a significant reduction of preoperative neurologic complication rates.4e6 The initiation of the best pharmacotherapy, including antiplatelet drugs and statins, and adequate blood pressure control immediately after the diagnosis of carotid atherosclerotic disease diminishes the risk of stroke in symptomatic patients waiting for surgery, during surgery, and after surgery.7e9 The beneficial role of CEA is currently supported by level 1 evidence in selected asymptomatic 901
902 Van den Brande et al.
and symptomatic patients with a low perioperative risk.10 Advances in surgery and anesthesiology have resulted in fewer postoperative intensive care unit admissions11 and in patient discharge as early as the evening of the day of surgery.12,13 However, when postoperative hospital length of stay is shortened, safety concerns require the timely preoperative identification of patients at risk, early recognition of postoperative complications, and a cautionary discharge policy to prevent readmissions for complications that could have been anticipated. Factors associated with a prolonged length of postoperative stay include age, insulin-dependent diabetes, female sex, chronic renal insufficiency, the need for intravenous vasodilators, undue delay of discharge, and postoperative complications, such as myocardial infarction, central neurologic deficits, and neck hematoma.13 Neurologic deficits and neck hematomas after CEA occur within 8 hours postoperatively in 95% of patients. After 24 hours, additional complications are infrequent and unpredictable.13 Discharge on the first postoperative day has been considered feasible in many patients undergoing CEA as long as the medical risk factors necessitating a prolonged postoperative length of stay have been identified.14,15 In our view, early discharge and returning home may also require the social and actual support of family or friends. In view of this, we conducted a prospective study to investigate the feasibility and safety of discharge on the first postoperative day after elective CEA and sought to identify the reasons for a prolonged length of stay and for readmission within 30 days.
METHODS Between June 2011 and January 2012, consecutive patients scheduled to undergo elective CEA were enrolled in this prospective study. The study protocol excluded patients undergoing CEA within 24 hours after hospital admission for ischemic neurologic deficit because we considered that this would not leave enough time to consult adequately with the family about the aim and the postoperative conditions of the study, patients having CEA together with heart surgery because this would prevent hospital discharge on the first postoperative day, and those undergoing carotid artery stenting. Indications for CEA included a degree of stenosis of at least 80% in asymptomatic patients and a degree of at least 50% in symptomatic patients, as determined by computed tomography angiography. Patients were considered symptomatic if they had a history of stroke, transient ischemic attack, or amaurosis fugax related to the carotid stenosis in the 6 months before surgery or when nonlateralizing symptoms existed
Annals of Vascular Surgery
that could be attributed to severe bilateral carotid stenosis. We introduced a uniform policy of discharge from the hospital on the first postoperative day after CEA. The patientsdand their relatives when presentdwere informed about this policy and were advised that discharge would occur on the first postoperative day only if there were no medical or surgical contraindications (i.e., medical criteria) and if the patient would not be left alone on the first day and night at home (i.e., social criteria). Patients living alone were advised to stay with family or friends during that period. In case of patients living in a home for the elderly or a center for physical rehabilitation, the management of the home was informed about the study. The local ethical vommittee approved the study. All patients enrolled in the study provided written informed consent. Preoperatively, risk factors for unfavorable outcome were identified and pharmacologic plaque stabilization and antiplatelet therapy were initiated. A thorough medical history, clinical examination, preoperative blood sampling, and analysis of medical records were performed to evaluate cardiovascular risk factors and comorbidities. Patients were considered hypertensive when diastolic pressure was >90 mm Hg or when they were using antihypertensive drugs. Hyperlipidemia was defined as a fasting serum concentration of cholesterol >190 mg/dL, low-density lipoprotein level of >115 mol/dL, triglycerides >180 mg/dL, or when patients were using antihyperlipidemic drugs. Patients were considered diabetic when they were taking oral antidiabetic medications or insulin or when the fasting serum glucose concentration was >125 mg/dL. Smokers were defined as patients who were currently smoking or who were smoking in the year before the study. Cardiac disease was defined present when an electrocardiogram showed signs of recent or remote myocardial infarction, asymptomatic arrhythmia, or when stable or unstable angina, congestive heart failure, or symptomatic arrhythmia existed clinically. Patients were considered to have pulmonary disease in the event of dyspnea on mild exertion or in cases of radiographic parenchymal changes or abnormal pulmonary function testing. Chronic renal insufficiency was defined as a serum creatinine concentration of >1.5 mg/dL or estimated glomerular filtration rate of <60 mL/min/1.73 m2. Perioperatively, all patients were taking statins, low-dose aspirin, and antihypertensive drugs when necessary. On the morning of the day of surgery, all oral medications prescribed, including antihypertensive drugs, were administered. The choice to discontinue clopidogrel was at the discretion of the surgeon.
Vol. 28, No. 4, May 2014
All patients underwent CEA under locoregional plexus anesthesia unless contraindications existed (i.e., impaired perioperative communication in case of foreign languageespeaking patients, repeat surgery, hostile neck, extreme obesity, and patients with a fear of locoregional anesthesia). Blood pressure was recorded continuously during and after the procedure using a radial artery catheter. Standard CEA consisted of bifurcation endarterectomy and Dacron patch (DuPont, Kinston, NC) angioplasty and was performed by 4 operators. An endoluminal shunt was used in all cases under general anesthesia and in select cases under locoregional plexus anesthesia. Wound drainage was at the discretion of the surgeon. Postoperatively, all patients were brought to the postanesthesia care unit for neurologic observation and blood pressure monitoring and to check for bleeding or neck swelling. On the morning of the first postoperative day, all patients were transferred to the vascular surgery ward, where they received instructions on how to use an automatic sphygmomanometer. Nursing staff checked autonomous locomotion, such as leaving the bed, walking, eating, and drinking as well as getting dressed without help. Emphasis was given to the correct self-administration of all medications prescribed and to the importance of blood pressure measurements. All patients were provided a sphygmomanometer to perform blood pressure measurements every morning and evening for 3 consecutive days after the operation. Finally, patients were offered the mobile phone number of the study nurse who was available 24 hours a day in case of an emergency or in case of remaining questions. When medical and social criteria were fulfilled, patients were allowed to leave the hospital after lunch on the first postoperative day. The study nurse made an assessment of all patients by telephone on postoperative days 2, 3, and 4. Patients were asked for signs of potential cerebral ischemia, cerebral hyperperfusion, and for blood pressure measurements and were checked for correct drug use. If there was a suspicion of cerebral ischemia, cerebral hyperperfusion, or abnormal blood pressure, the family doctor was contacted to visit the patient. Finally, a home visit by the study nurse was performed in all patients on postoperative day 30 to record hospital readmissions, if any, and to retrieve the sphygmomanometer. Statistical Analysis The initial analyses of our data were exploratory and descriptive, with categorical data presented as numbers of patients and percentages and continuous data presented as means with standard deviations (SDs).
Discharge on postoperative day 1 after CEA 903
As a second step, the group of patients that was discharged from the hospital on the first postoperative day and the group of patients discharged at a later stage were compared using either the Fisher’s exact test (for binary categorical variables) or the chi-squared test (for categorical variables with >2 levels). In all cases, an unadjusted (crude) odds ratio (OR) was calculated using univariate logistic regression. In addition, ORs simultaneously adjusted for the factors that were found to be statistically significant in univariate analyses were calculated using multivariate regression analysis. Each OR is presented with its 95% confidence interval. All computational procedures were performed using Excel 2003 (Microsoft, Redmond, WA) and IBM SPSS software (version 20; IBM Corporation, Chicago, IL).
RESULTS During the enrollment period, all 58 patients scheduled for CEA were asked to participate in the study. All 58 consecutive patients provided informed consent and were initially included. However, 1 patient withdrew from the study before the day of surgery, after she had further discussed the study with relatives other than those who agreed earlier to stay with the patient at home after hospital discharge. Therefore, of the 58 patients asked to participate, 57 (98 %) were included. During the period of enrollment, no patients underwent CEA within 24 hours of the neurologic event, no patients underwent CEA together with heart surgery, and 2 patients had carotid artery stenting. Finally, 62 elective CEA procedures were carried out in 57 patients, because 5 patients were operated bilaterally. Table I shows the patient demographics, cardiovascular risk factors, comorbidities, and indications for CEA. The mean time period from last neurologic symptoms to CEA was 21 days (SD: 15.9). Eighteen of 35 (51%) symptomatic cases were operated within 2 weeks of the neurologic event. Fifty-three CEA procedures (85%) were carried out under locoregional anesthesia and 9 (14%) under general anesthesia. The standard technique of endarterectomy and angioplasty with a Dacron patch (DuPont) was used in 57 (92%) of the 62 CEAs. In 5 CEAs (8%), the carotid artery was closed with a simple running suture. There were no cases of in-patient perioperative stroke or death. Intravenous antihypertensive drugs to correct intraoperative and/or postoperative arterial hypertension were needed in 6 CEAs (10%). In these patients, treatment was discontinued before their transfer to the surgical ward after blood pressure normalization. In 9 CEAs (15%) a visible neck hematoma developed and was treated
904 Van den Brande et al.
Annals of Vascular Surgery
Table I. Demographics, cardiovascular risk factors, and comorbidities in 57 patients undergoing 62 carotid endarterectomy procedures and indications for carotid endarterectomy Demographics, cardiovascular risk factors, and comorbidities
Mean age, yr (range) Men, n (%) Women, n (%) Hypertension, n (%) Hyperlipidemia, n (%) Diabetes, n (%) Smoking, n (%) Cardiac disease, n (%) Pulmonary disease, n (%) Chronic renal insufficiency, n (%) Indications for CEA (n ¼ 62) Asymptomatic 80% stenosis, n (%) Transient ischaemic attack, n (%) Amaurosis fugax, n (%) Recovered stroke, n (%) Nonlateralizing symptoms, n (%)
N ¼ 57
74.2 33 24 48 42 22 16 40 18 6 27 12 6 16 1
(51e89) (58) (42) (84) (74) (39) (28) (70) (32) (11) (44) (19) (10) (26) (2)
CEA, carotid endarterectomy.
Table II. Day of discharge and medical and social reasons for discharge later than postoperative day 1 in 62 carotid endarterectomy procedures
Day 1 Later than day 1 Day 2 Day 3 Day 10
Medicala
Socialb
n (%)
n (%)
n (%)
45 17 15 1 1
d 5 (9) 3 (5) 1 (2) 1 (2)
d 12 (19) 12 (19)c d d
(73) (27) (24) (2) (2)
a
Postoperative day 2: planned pacemaker check-up in 1 CEA patient and planned check-up for aortic valve dysfunction in 2 CEA patients; postoperative day 3: episode of confusion, without diagnosis, in 1 CEA patient; and postoperative day 10: extreme obesity and further check-up for hypoventilation syndrome in 1 CEA patient. b CEA patient living alone and no relative or friend available to stay with the patient for the first day and night after intentional discharge on the first postoperative day. c Twelve CEAs in female patients.
conservatively in all cases. The medical and social criteria to be discharged from the hospital on the first postoperative day were fulfilled in 45 (72%) of the 62 CEA procedures, and all of these patients were effectively discharged. Discharge from the hospital was postponed to the second postoperative day or later in 17 (27%) of the CEA procedures (Table II). All 12 patients discharged on the second postoperative day had this delayed discharge
because of social reasons: all of the latter were older female patients (mean age: 75 years) who lived alone and who had no family or friends available to stay with them during the first day and night after an intended discharge on the first postoperative day. Table III shows patient demographics, cardiovascular risk factors, comorbidities, and type of anesthesia related to the day of discharge. Statistically significant risk factors for prolonged length of postoperative hospital stay were female sex, the use of general anesthesia, and the presence of chronic renal insufficiency, both in univariate (i.e., considering the effect of each characteristic separately) and multivariate analyses (i.e., considering the simultaneous effect of female sex, the presence or absence of renal disease, and type of anesthesia). We found no relationship between the length of postoperative stay and age, indication for surgery, side of surgery, smoking status, and the presence of diabetes, hyperlipidemia, arterial hypertension, and cardiac or pulmonary disease. Also, the time period from last neurologic symptoms to CEA was not associated with length of stay after CEA, with a mean time period from last symptoms to CEA of 23 days (SD: 14.9) and 19 days (SD: 17.4) for patients staying >1 day after CEA, respectively (P ¼ 0.43). Postoperative follow-up of 30 days was completed in all cases. In this postoperative period, no transient ischemic symptoms or genuine ischemic strokes occurred. Eight patients (representing 13% of the CEA procedures) were readmitted to the hospital during this 30-day postoperative period. However, no readmission was related to the CEA procedure itself. Four of these readmissions were planned readmissions for additional elective vascular surgical procedures, 2 were because of trauma, and 2 were for medical reasons (Table IV).
DISCUSSION In this clinical series of 62 consecutive cases of CEA with uneventful neurologic postoperative outcome in 57 patients, patients were discharged safely on the first postoperative day in 73% of CEA cases. There were no cases of perioperative stroke or death within a 30-day follow-up period. Despite this early discharge, there were no hospital readmissions related to CEA in the follow-up period. This good surgical outcome may be partly attributed to the surgical experience of the operators and their attention to potential risk factors for postoperative complications,16e18 to the implementation of the best possible preoperative and
Vol. 28, No. 4, May 2014
Discharge on postoperative day 1 after CEA 905
Table III. Patient demographics, cardiovascular risk factors, comorbidities, and type of anesthesia used related to the day of discharge
Characteristic
Age >75 yrs, n (%) Female sex, n (%) Indication for surgery Symptomatic stenosis Left-sided CEA Arterial hypertension Hyperlipidemia Comorbidity Diabetes Tobacco Cardiac disease Pulmonary disease Chronic renal insufficiency General anesthesia
Discharge on postoperative day 1 (n ¼ 45)
Discharge on postoperative day 2 or later (n ¼ 17)
P valuea
21 (47%) 14 (31%)
12 (71%) 12 (71%)
0.153 0.009
2.74 (0.83e9.07) 5.32 (1.57e17.99)
d 4.70 (1.16e19.00)
22 22 37 34
(49%) (49%) (82%) (76%)
13 6 16 12
(76%) (35%) (94%) (71%)
0.084 0.400 0.423 0.750
3.40 0.57 3.46 0.78
(0.86e16.22) (0.18e1.81) (0.40e29.99) (0.22e2.69)
d d d d
13 20 28 13 1
(29%) (44%) (62%) (29%) (2%)
5 4 12 4 5
(29%) (24%) (71%) (24%) (29%)
1.000 0.156 0.767 0.759 0.005
1.03 0.39 1.46 0.76 18.33
(0.30e3.49) (0.11e1.36) (0.44e4.86) (0.21e2.76) (1.95e172.20)
d d d d 22.71 (1.70e302.81)
10 (22.2%)
10 (58.8%)
0.013
Unadjusted OR (95% CI)b
5.00 (1.51e16.51)
Adjusted OR (95% CI)c
4.91 (1.21e19.95)
CEA, carotid endarterectomy; CI, confidence interval; OR, odds ratio. a P value for between-group difference calculated using the Fisher’s exact or chi-squared tests. P < 0.05 is considered statistically significant. b Crude OR (95% CI) calculated using univariate logistic regression (i.e., considering the effect of each characteristic separately). c OR simultaneously adjusted for the factors that were found to be statistically significant in univariate analyses (i.e., the current adjusted, multivariate analysis considers the effect of female sex, the presence or absence of renal disease, and type of anesthesia simultaneously).
Table IV. Day and reason for hospital readmission of 8 patients in the 30-day postoperative period after 62 carotid endarterectomy procedures Reason
Other vascular surgery Endoprosthesis for abdominal aneurysm Contralateral carotid endarterectomy Femoral popliteal bypass procedure Open abdominal aneurysm repair Trauma Shoulder dislocation Severe hand trauma Medical Pneumonia Gastrointestinal bleeding and acenocoumarol intoxication
Postoperative day
7 30 30 30 21 21 14 30
intraoperative medical treatment,7,8 and finally to the extensive preoperative counseling of patients and their family with regard to early postoperative discharge. The latter practice is highly demanding because intentional early discharge after this type of delicate surgery requires a full explanation of
the perioperative risks, a high level of confidence between the physician and the patient and his/her family, and the availability of the patient’s relatives or friends in spite of potential interference with the activities of the latter. In addition, an efficient early discharge policy relies on full cooperation between the surgeon and the ward nursing staff to abide by the early discharge protocol to prevent undue delay of discharge. Kaufman et al.19 have previously investigated the feasibility and safety of discharge on the first postoperative day after CEA in a cohort of asymptomatic and symptomatic patients comparable to our study population. These authors achieved this goal in 60% of their patients. However, 13% of the patients in that study experienced complications within 30 days after CEA, including death in 3 patients and neurologic deficits in 5 patients. The poorer outcomes documents in this study, conducted >15 years ago, compared to the study presented here might be explained by an inferior medical therapy at that time. Nevertheless, Kaufman et al.19 also concluded that a short postoperative length of stay was not associated with a significant risk of unplanned readmission. According to the data of our prospective study, factors predisposing to postponed discharge were female sex, chronic renal insufficiency, and general
906 Van den Brande et al.
anesthesia. Postponed discharge of women was not attributable to medical reasons, nor to cardiovascular risk factors or comorbidities in any of the cases, but to the requirement of the study protocol to have someone available to stay with the patient after discharge; elderly patients were more frequently widowed and living alone than male patients. Collier20 reported that a clinical pathway including extensive preoperative patient education and a streamlined postoperative patient care permitted safe discharge on the first postoperative day in 84% of patients who had undergone CEA. This percentage is higher than that achieved in our study, but Collier’s study protocol20 did not require a relative to stay with the patient after discharge. It is possible that by omitting this social criterion the number of patients who can be discharged on the first postoperative day may increase. The identification of chronic renal insufficiency as a significant risk factor for a longer postoperative length of stay in our study has previously been reported.13,21 General anesthesia was the third factor associated with a prolonged postoperative stay in our study. Earlier studies have shown a reduced length of stay after CEA performed under local or regional anaesthesia.22e24 The superior postoperative neurocognitive outcome after CEA performed under local anesthesia may facilitate a quicker recovery and therefore earlier discharge.25 Other risk factors, including comorbidities and postoperative complications, such as preoperative neurologic symptoms, older age, diabetes mellitus, requirement of intravenous vasodilators, or postoperative neurologic or surgical morbidity14,15,21,26,27 that may preclude safe and early discharge after CEA were not significantly related to hospital length of stay in our study. None of the events of readmission to the hospital as recorded during the 30-day postoperative period were attributable to early discharge, which is in line with previous studies.19,21,28 Our present study concerns only a limited number of patients undergoing CEA, but is prospective and confirms most findings of former, mostly retrospective studies. A historical comparison in our department of length of postoperative stay before and during the present study was not attempted because before the study, patients were discharged at the discretion of the operating or ward surgeon, based only on clinical arguments. What is new here is that we looked for social issues that might render a safe and early discharge as inappropriate. Now, after the present study, we preoperatively propose this first postoperative day discharge to the patient if medically possible. However, we strongly advise older patients to have a relative to be with
Annals of Vascular Surgery
them for 24 hours. It was not the purpose of the study (or our intent) to calculate or to have impact on reimbursements or hospital resources. The goal was to increase patient satisfaction by early discharge but to be as safe as possible while doing so. Of special concern must be cerebral hyperperfusion syndrome. It is a relatively rare, but potentially devastating event that can occur a brief time after CEA. It is difficult to predict, probably because of increased arterial blood pressure, and occurs mostly in cases of nonelective CEA. Although the definite diagnosis has to be confirmed by imaging techniques, the presence of clinical symptoms is essential: severe headache, ocular or facial pain, confusion, visual disturbances, epileptic seizures, or any focal deficits not caused by cerebral ischemia. Awareness for these clinical signs and control of blood pressure are both part of postoperative patient surveillance. When present or when blood pressure is poorly controlled, early discharge is of course precluded.30,31 In the continuous evolution toward shortening of the perioperative length of stay for CEA, 1-day admission20 and same-evening discharge13,29 policies have also been adopted and have been shown to be feasible, but we believe that these policies may surpass the limits of safe surgical practice. CEA remains a controversial vascular surgical procedure, and any additional approach that could potentially reduce its efficacy should be viewed with extreme caution.32
CONCLUSION Discharge on the first postoperative day after elective CEA can be a safe and feasible policy in many patients, and unplanned readmissions linked to this early discharge must be uncommon. The social environment of the patient appeared to be the major factor preventing early discharge in our study. Appropriate preoperative patient and family counseling is essential before implementing an early discharge policy. Finally, in the continuous evolution of perioperative care of patients undergoing CEA, preoperative diagnostic work-up and risk factor management, perioperative antiplatelet therapy, experienced surgical and anesthetic practice, and excellent postoperative care must and will remain a real challenge in the additional reduction of perioperative neurologic morbidity and mortality. REFERENCES 1. North American Symptomatic Carotid Endarterectomy Trial Collaborators. Beneficial effect of carotid endarterectomy in symptomatic patients with high grade stenosis. N Engl Med 1991;325:445e8.
Vol. 28, No. 4, May 2014
2. Executive Committee for the Asymptomatic Carotid Atherosclerosis Study (ACAS). Endarterectomy for asymptomatic carotid artery stenosis. JAMA 1999;273:1424e8. 3. European Carotid Surgery Trialists’ Collaborative Group. Randomised trial of endarterectomy for recently symptomatic carotid stenosis: final results of the MRC European Carotid Surgery Trial (ECST). Lancet 1998;351:1379e87. 4. Berczi V, Randall M, Balamurugan R, et al. Safety of arch aortography for assessment of carotid arteries. Eur J Vasc Endovasc Surg 2006;31:3e7. 5. Borisch I, Horn M, Butz B, et al. Preoperative evaluation of carotid artery stenosis: comparison of contrast-enhanced MR angiography and duplex sonography with digital subtraction angiography. Am J Neuroradiol 2003;24:1117e22. 6. Johnston DC, Eastwood JD, Nguyen T, et al. Contrastenhanced magnetic resonance angiography of carotid arteries. Utility in routine clinical practice. Stroke 2002;33:2834e8. 7. Naylor AR. The importance of initiating ‘‘best medical therapy’’ and intervening as soon as possible in patients with symptomatic carotid artery disease: time for a radical rethink of practice. J Cardiovasc Surg 2009;50:773e82. 8. McGirt MJ, Perler BA, Brooke BS, et al. 3-hydroxy-3methylglutaryl coenzyme A reductase inhibitors reduce the risk of perioperative stroke and mortality after carotid endarterectomy. J Vasc Surg 2005;42:829e36. 9. Peeters W, Hellings WE, de Kleijn DP, et al. Carotid atherosclerotic plaques stabilize after stroke: insights into the natural process of atherosclerotic plaque stabilization. Arterioscler Thromb Vasc Biol 2009;29:128e33. 10. Liapis CD, Bell PRF, Mikhailidis D, et al. ESVS guidelines. Invasive treatment for carotid stenosis: indications and techniques. Eur J Vasc Endovasc Surg 2009;37(suppl 1):S1e19. 11. Hobart DC, Nicholas GG, Reed JF 3rd, et al. Carotid endarterectomy outcomes research: reduced resource utilization using a clinical protocol. Cardiovasc Surg 2000;8:446e51. 12. Angevine PD, Choudhri TF, Huang J, et al. Significant reductions in length of stay after carotid endarterectomy can be safely accomplished without modifying either anesthetic technique or postoperative ICU monitoring. Stroke 1999;30:2341e6. 13. Sheehan MK, Baker WH, Littooy FN, et al. Timing of postcarotid complications: a guide to safe discharge planning. J Vasc Surg 2001;34:13e6. 14. Roddy S, Estes J, Kwoun M, et al. Factors predicting prolonged length of stay after carotid endarterectomy. J Vasc Surg 2000;32:550e4. 15. Hernandez N, Salles-Cunha SX, Daoud YA, et al. Factors related to short length of stay after carotid endarterectomy. Vasc Endovasc Surg 2002;36:425e37. 16. Debing E, Aerden D, Van den Brande P. Diabetes mellitus is a predictor for early adverse outcome after carotid endarterectomy. Vasc Endovasc Surg 2011;45:28e32.
Discharge on postoperative day 1 after CEA 907
17. Debing E, Van den Brande P. Chronic renal insufficiency and risk of early mortality in patients undergoing carotid endarterectomy. Ann Vasc Surg 2006;20:609e13. 18. Debing E, Van den Brande P. Carotid endarterectomy in the elderly: are the patient characteristics, the early outcome, and the predictors the same as those in younger patients. Surg Neurol 2007;67:467e71. 19. Kaufman JL, Frank D, Rhee SW, et al. Feasibility and safety of 1-day postoperative hospitalization for carotid endarterectomy. Ann Surg 1996;131:751e5. 20. Collier PE. Are one-day admissions for carotid endarterectomy feasible? Am J Surg 1995;170:140e3. 21. Cikrit DF, Larson DM, Sawchuk AP, et al. Short- stay carotid endarterectomy in a tertiary-care Veterans Administration hospital. Am J Surg 2004;188:544e8. 22. McCarthy RJ, Walker R, McAteer P, et al. Patient and hospital benefits of local anaesthesia for carotid endarterectomy. Eur J Vasc Endovasc Surg 2001;22:13e8. 23. Mofidi R, Nimmo AF, Moores C, et al. Regional versus general anaesthesia for carotid endarterectomy: impact of change in practice. Surgeon 2006;4:158e62. 24. Kalko Y, Kafali E, Aydin U, et al. Surgery of the carotid artery: local anaesthesia versus general anaesthesia. Acta Chir Belg 2007;107:53e7. 25. Weber CF, Friedl H, Hueppe M, et al. Impact of general versus local anaesthesia on early postoperative cognitive dysfunction following carotid endarterectomy: GALA Study Subgroup Analysis. World J Surg 2009;33: 1526e32. 26. Katz SG, Kohl RD. Carotid endarterectomy with shortened hospital stay. Arch Surg 1995;130:887e90. 27. Young KC, Jahromi BS, Singh MJ, et al. Hospital resource use following carotid endarterectomy in 2006: analysis of the nationwide inpatient sample. J Stroke Cerebrovasc Dis 2010;19:458e64. 28. Su LT, Carpenter JP. Decreasing carotid endarterectomy length of stay at a university hospital. Cardiovasc Surg 1999;7:292e7. 29. Sheehan MK, Greisler HP, Littooy FN, et al. Same-evening discharge after carotid endarterectomy: our initial experience. J Vasc Surg 2004;39:575e7. 30. De Rango P. Cerebral hyperperfusion syndrome: the dark side of carotid endarterectomy. Eur J Vasc Endovasc Surg 2012;43:377. 31. Maas MB, Kwolek CJ, Hirsch JA, et al. Clinical risk predictors for cerebral hyperperfusion syndrome after carotid endarterectomy. J Neurol Neurosurg Psychiatry 2013;84: 569e72. 32. Ricotta JJ. Regarding: ‘‘Timing of postcarotid complications: a guide to safe discharge planning’’. J Vasc Surg 2001;34: 178e9.
Background: Medical complications may prolong the hospital stay after elective carotid endarterectomy (CEA). We prospectively assessed the social and medical feasibility and safety of patient discharge on the first postoperative day after elective CEA and unplanned readmissions. Methods: Between June 2011 and January 2012, 57 consecutive patients scheduled for elective CEA were enrolled with the aim of discharge on the first postoperative day if there were no medical contraindications and on the condition that the patient should not be left alone during the first day and night at home. CEA was carried out under local or general anesthesia. After discharge, the patients were contacted to ascertain the occurrence of arterial hypertension, cerebral hyperperfusion, focal cerebral ischemia, or hospital readmission. Results: Sixty-two CEA were carried out in 57 patients (33 men and 24 women ranging in age from 51e89 years). The indications for CEA were: asymptomatic high grade stenosis in 27, hemispheric transient ischemic attack in 12, amaurosis fugax in 6, recovered stroke in 16, and nonlateralizing signs in 1. There were no cases of perioperative stroke or death. Discharge on the first postoperative day was achieved in 45 cases (73%). In 15 cases (24%), discharge was on the second postoperative day because of the absence of a relative (12 cases) or for medical reasons (3 cases). Discharge was on day 3 in 1 case, and on day 10 in another, both for medical reasons. No cases of severe arterial hypertension, stroke, mortality, or readmission for reasons related to the CEA procedure were recorded up to postoperative day 30. Conclusion: In this study, the majority of patients undergoing elective CEA were discharged safely on the first postoperative day. Social reasons, rather than medical reasons, underlied most cases of later discharge. There were no unplanned readmissions for complications of CEA.
INTRODUCTION Consistent with The North American Symptomatic Carotid Endarterectomy Trial,1 the Asymptomatic Carotid Artery Study,2 and the European Carotid 1 Department of Vascular Surgery, Universitair Ziekenhuis Brussel, Brussels, Belgium. 2 Department of Anesthesiology, Universitair Ziekenhuis Brussel, Brussels, Belgium. 3 Center for Outcome Research, Universitair Ziekenhuis Brussel, Brussels, Belgium.
Correspondence to: Pierre Van den Brande, MD, PhD, Universitair Ziekenhuis Brussel, Department of Vascular Surgery, Laarbeeklaan 101, B-1090 Brussels, Belgium; E-mail: [email protected] Ann Vasc Surg 2014; 28: 901–907 http://dx.doi.org/10.1016/j.avsg.2013.10.014 Ó 2014 Elsevier Inc. All rights reserved. Manuscript received: July 8, 2013; manuscript accepted: October 6, 2013.
Surgery Trial,3 carotid endarterectomy (CEA) has been proposed as a treatment option for carotid atherosclerotic disease in proportion to the degree of stenosis because of its beneficial role in the prevention of stroke. The replacement of selective catheter angiography for diagnostic assessment by noninvasive imaging approaches, such as magnetic resonance angiography, computed tomography angiography, and Doppler ultrasonography have resulted in a significant reduction of preoperative neurologic complication rates.4e6 The initiation of the best pharmacotherapy, including antiplatelet drugs and statins, and adequate blood pressure control immediately after the diagnosis of carotid atherosclerotic disease diminishes the risk of stroke in symptomatic patients waiting for surgery, during surgery, and after surgery.7e9 The beneficial role of CEA is currently supported by level 1 evidence in selected asymptomatic 901
902 Van den Brande et al.
and symptomatic patients with a low perioperative risk.10 Advances in surgery and anesthesiology have resulted in fewer postoperative intensive care unit admissions11 and in patient discharge as early as the evening of the day of surgery.12,13 However, when postoperative hospital length of stay is shortened, safety concerns require the timely preoperative identification of patients at risk, early recognition of postoperative complications, and a cautionary discharge policy to prevent readmissions for complications that could have been anticipated. Factors associated with a prolonged length of postoperative stay include age, insulin-dependent diabetes, female sex, chronic renal insufficiency, the need for intravenous vasodilators, undue delay of discharge, and postoperative complications, such as myocardial infarction, central neurologic deficits, and neck hematoma.13 Neurologic deficits and neck hematomas after CEA occur within 8 hours postoperatively in 95% of patients. After 24 hours, additional complications are infrequent and unpredictable.13 Discharge on the first postoperative day has been considered feasible in many patients undergoing CEA as long as the medical risk factors necessitating a prolonged postoperative length of stay have been identified.14,15 In our view, early discharge and returning home may also require the social and actual support of family or friends. In view of this, we conducted a prospective study to investigate the feasibility and safety of discharge on the first postoperative day after elective CEA and sought to identify the reasons for a prolonged length of stay and for readmission within 30 days.
METHODS Between June 2011 and January 2012, consecutive patients scheduled to undergo elective CEA were enrolled in this prospective study. The study protocol excluded patients undergoing CEA within 24 hours after hospital admission for ischemic neurologic deficit because we considered that this would not leave enough time to consult adequately with the family about the aim and the postoperative conditions of the study, patients having CEA together with heart surgery because this would prevent hospital discharge on the first postoperative day, and those undergoing carotid artery stenting. Indications for CEA included a degree of stenosis of at least 80% in asymptomatic patients and a degree of at least 50% in symptomatic patients, as determined by computed tomography angiography. Patients were considered symptomatic if they had a history of stroke, transient ischemic attack, or amaurosis fugax related to the carotid stenosis in the 6 months before surgery or when nonlateralizing symptoms existed
Annals of Vascular Surgery
that could be attributed to severe bilateral carotid stenosis. We introduced a uniform policy of discharge from the hospital on the first postoperative day after CEA. The patientsdand their relatives when presentdwere informed about this policy and were advised that discharge would occur on the first postoperative day only if there were no medical or surgical contraindications (i.e., medical criteria) and if the patient would not be left alone on the first day and night at home (i.e., social criteria). Patients living alone were advised to stay with family or friends during that period. In case of patients living in a home for the elderly or a center for physical rehabilitation, the management of the home was informed about the study. The local ethical vommittee approved the study. All patients enrolled in the study provided written informed consent. Preoperatively, risk factors for unfavorable outcome were identified and pharmacologic plaque stabilization and antiplatelet therapy were initiated. A thorough medical history, clinical examination, preoperative blood sampling, and analysis of medical records were performed to evaluate cardiovascular risk factors and comorbidities. Patients were considered hypertensive when diastolic pressure was >90 mm Hg or when they were using antihypertensive drugs. Hyperlipidemia was defined as a fasting serum concentration of cholesterol >190 mg/dL, low-density lipoprotein level of >115 mol/dL, triglycerides >180 mg/dL, or when patients were using antihyperlipidemic drugs. Patients were considered diabetic when they were taking oral antidiabetic medications or insulin or when the fasting serum glucose concentration was >125 mg/dL. Smokers were defined as patients who were currently smoking or who were smoking in the year before the study. Cardiac disease was defined present when an electrocardiogram showed signs of recent or remote myocardial infarction, asymptomatic arrhythmia, or when stable or unstable angina, congestive heart failure, or symptomatic arrhythmia existed clinically. Patients were considered to have pulmonary disease in the event of dyspnea on mild exertion or in cases of radiographic parenchymal changes or abnormal pulmonary function testing. Chronic renal insufficiency was defined as a serum creatinine concentration of >1.5 mg/dL or estimated glomerular filtration rate of <60 mL/min/1.73 m2. Perioperatively, all patients were taking statins, low-dose aspirin, and antihypertensive drugs when necessary. On the morning of the day of surgery, all oral medications prescribed, including antihypertensive drugs, were administered. The choice to discontinue clopidogrel was at the discretion of the surgeon.
Vol. 28, No. 4, May 2014
All patients underwent CEA under locoregional plexus anesthesia unless contraindications existed (i.e., impaired perioperative communication in case of foreign languageespeaking patients, repeat surgery, hostile neck, extreme obesity, and patients with a fear of locoregional anesthesia). Blood pressure was recorded continuously during and after the procedure using a radial artery catheter. Standard CEA consisted of bifurcation endarterectomy and Dacron patch (DuPont, Kinston, NC) angioplasty and was performed by 4 operators. An endoluminal shunt was used in all cases under general anesthesia and in select cases under locoregional plexus anesthesia. Wound drainage was at the discretion of the surgeon. Postoperatively, all patients were brought to the postanesthesia care unit for neurologic observation and blood pressure monitoring and to check for bleeding or neck swelling. On the morning of the first postoperative day, all patients were transferred to the vascular surgery ward, where they received instructions on how to use an automatic sphygmomanometer. Nursing staff checked autonomous locomotion, such as leaving the bed, walking, eating, and drinking as well as getting dressed without help. Emphasis was given to the correct self-administration of all medications prescribed and to the importance of blood pressure measurements. All patients were provided a sphygmomanometer to perform blood pressure measurements every morning and evening for 3 consecutive days after the operation. Finally, patients were offered the mobile phone number of the study nurse who was available 24 hours a day in case of an emergency or in case of remaining questions. When medical and social criteria were fulfilled, patients were allowed to leave the hospital after lunch on the first postoperative day. The study nurse made an assessment of all patients by telephone on postoperative days 2, 3, and 4. Patients were asked for signs of potential cerebral ischemia, cerebral hyperperfusion, and for blood pressure measurements and were checked for correct drug use. If there was a suspicion of cerebral ischemia, cerebral hyperperfusion, or abnormal blood pressure, the family doctor was contacted to visit the patient. Finally, a home visit by the study nurse was performed in all patients on postoperative day 30 to record hospital readmissions, if any, and to retrieve the sphygmomanometer. Statistical Analysis The initial analyses of our data were exploratory and descriptive, with categorical data presented as numbers of patients and percentages and continuous data presented as means with standard deviations (SDs).
Discharge on postoperative day 1 after CEA 903
As a second step, the group of patients that was discharged from the hospital on the first postoperative day and the group of patients discharged at a later stage were compared using either the Fisher’s exact test (for binary categorical variables) or the chi-squared test (for categorical variables with >2 levels). In all cases, an unadjusted (crude) odds ratio (OR) was calculated using univariate logistic regression. In addition, ORs simultaneously adjusted for the factors that were found to be statistically significant in univariate analyses were calculated using multivariate regression analysis. Each OR is presented with its 95% confidence interval. All computational procedures were performed using Excel 2003 (Microsoft, Redmond, WA) and IBM SPSS software (version 20; IBM Corporation, Chicago, IL).
RESULTS During the enrollment period, all 58 patients scheduled for CEA were asked to participate in the study. All 58 consecutive patients provided informed consent and were initially included. However, 1 patient withdrew from the study before the day of surgery, after she had further discussed the study with relatives other than those who agreed earlier to stay with the patient at home after hospital discharge. Therefore, of the 58 patients asked to participate, 57 (98 %) were included. During the period of enrollment, no patients underwent CEA within 24 hours of the neurologic event, no patients underwent CEA together with heart surgery, and 2 patients had carotid artery stenting. Finally, 62 elective CEA procedures were carried out in 57 patients, because 5 patients were operated bilaterally. Table I shows the patient demographics, cardiovascular risk factors, comorbidities, and indications for CEA. The mean time period from last neurologic symptoms to CEA was 21 days (SD: 15.9). Eighteen of 35 (51%) symptomatic cases were operated within 2 weeks of the neurologic event. Fifty-three CEA procedures (85%) were carried out under locoregional anesthesia and 9 (14%) under general anesthesia. The standard technique of endarterectomy and angioplasty with a Dacron patch (DuPont) was used in 57 (92%) of the 62 CEAs. In 5 CEAs (8%), the carotid artery was closed with a simple running suture. There were no cases of in-patient perioperative stroke or death. Intravenous antihypertensive drugs to correct intraoperative and/or postoperative arterial hypertension were needed in 6 CEAs (10%). In these patients, treatment was discontinued before their transfer to the surgical ward after blood pressure normalization. In 9 CEAs (15%) a visible neck hematoma developed and was treated
904 Van den Brande et al.
Annals of Vascular Surgery
Table I. Demographics, cardiovascular risk factors, and comorbidities in 57 patients undergoing 62 carotid endarterectomy procedures and indications for carotid endarterectomy Demographics, cardiovascular risk factors, and comorbidities
Mean age, yr (range) Men, n (%) Women, n (%) Hypertension, n (%) Hyperlipidemia, n (%) Diabetes, n (%) Smoking, n (%) Cardiac disease, n (%) Pulmonary disease, n (%) Chronic renal insufficiency, n (%) Indications for CEA (n ¼ 62) Asymptomatic 80% stenosis, n (%) Transient ischaemic attack, n (%) Amaurosis fugax, n (%) Recovered stroke, n (%) Nonlateralizing symptoms, n (%)
N ¼ 57
74.2 33 24 48 42 22 16 40 18 6 27 12 6 16 1
(51e89) (58) (42) (84) (74) (39) (28) (70) (32) (11) (44) (19) (10) (26) (2)
CEA, carotid endarterectomy.
Table II. Day of discharge and medical and social reasons for discharge later than postoperative day 1 in 62 carotid endarterectomy procedures
Day 1 Later than day 1 Day 2 Day 3 Day 10
Medicala
Socialb
n (%)
n (%)
n (%)
45 17 15 1 1
d 5 (9) 3 (5) 1 (2) 1 (2)
d 12 (19) 12 (19)c d d
(73) (27) (24) (2) (2)
a
Postoperative day 2: planned pacemaker check-up in 1 CEA patient and planned check-up for aortic valve dysfunction in 2 CEA patients; postoperative day 3: episode of confusion, without diagnosis, in 1 CEA patient; and postoperative day 10: extreme obesity and further check-up for hypoventilation syndrome in 1 CEA patient. b CEA patient living alone and no relative or friend available to stay with the patient for the first day and night after intentional discharge on the first postoperative day. c Twelve CEAs in female patients.
conservatively in all cases. The medical and social criteria to be discharged from the hospital on the first postoperative day were fulfilled in 45 (72%) of the 62 CEA procedures, and all of these patients were effectively discharged. Discharge from the hospital was postponed to the second postoperative day or later in 17 (27%) of the CEA procedures (Table II). All 12 patients discharged on the second postoperative day had this delayed discharge
because of social reasons: all of the latter were older female patients (mean age: 75 years) who lived alone and who had no family or friends available to stay with them during the first day and night after an intended discharge on the first postoperative day. Table III shows patient demographics, cardiovascular risk factors, comorbidities, and type of anesthesia related to the day of discharge. Statistically significant risk factors for prolonged length of postoperative hospital stay were female sex, the use of general anesthesia, and the presence of chronic renal insufficiency, both in univariate (i.e., considering the effect of each characteristic separately) and multivariate analyses (i.e., considering the simultaneous effect of female sex, the presence or absence of renal disease, and type of anesthesia). We found no relationship between the length of postoperative stay and age, indication for surgery, side of surgery, smoking status, and the presence of diabetes, hyperlipidemia, arterial hypertension, and cardiac or pulmonary disease. Also, the time period from last neurologic symptoms to CEA was not associated with length of stay after CEA, with a mean time period from last symptoms to CEA of 23 days (SD: 14.9) and 19 days (SD: 17.4) for patients staying >1 day after CEA, respectively (P ¼ 0.43). Postoperative follow-up of 30 days was completed in all cases. In this postoperative period, no transient ischemic symptoms or genuine ischemic strokes occurred. Eight patients (representing 13% of the CEA procedures) were readmitted to the hospital during this 30-day postoperative period. However, no readmission was related to the CEA procedure itself. Four of these readmissions were planned readmissions for additional elective vascular surgical procedures, 2 were because of trauma, and 2 were for medical reasons (Table IV).
DISCUSSION In this clinical series of 62 consecutive cases of CEA with uneventful neurologic postoperative outcome in 57 patients, patients were discharged safely on the first postoperative day in 73% of CEA cases. There were no cases of perioperative stroke or death within a 30-day follow-up period. Despite this early discharge, there were no hospital readmissions related to CEA in the follow-up period. This good surgical outcome may be partly attributed to the surgical experience of the operators and their attention to potential risk factors for postoperative complications,16e18 to the implementation of the best possible preoperative and
Vol. 28, No. 4, May 2014
Discharge on postoperative day 1 after CEA 905
Table III. Patient demographics, cardiovascular risk factors, comorbidities, and type of anesthesia used related to the day of discharge
Characteristic
Age >75 yrs, n (%) Female sex, n (%) Indication for surgery Symptomatic stenosis Left-sided CEA Arterial hypertension Hyperlipidemia Comorbidity Diabetes Tobacco Cardiac disease Pulmonary disease Chronic renal insufficiency General anesthesia
Discharge on postoperative day 1 (n ¼ 45)
Discharge on postoperative day 2 or later (n ¼ 17)
P valuea
21 (47%) 14 (31%)
12 (71%) 12 (71%)
0.153 0.009
2.74 (0.83e9.07) 5.32 (1.57e17.99)
d 4.70 (1.16e19.00)
22 22 37 34
(49%) (49%) (82%) (76%)
13 6 16 12
(76%) (35%) (94%) (71%)
0.084 0.400 0.423 0.750
3.40 0.57 3.46 0.78
(0.86e16.22) (0.18e1.81) (0.40e29.99) (0.22e2.69)
d d d d
13 20 28 13 1
(29%) (44%) (62%) (29%) (2%)
5 4 12 4 5
(29%) (24%) (71%) (24%) (29%)
1.000 0.156 0.767 0.759 0.005
1.03 0.39 1.46 0.76 18.33
(0.30e3.49) (0.11e1.36) (0.44e4.86) (0.21e2.76) (1.95e172.20)
d d d d 22.71 (1.70e302.81)
10 (22.2%)
10 (58.8%)
0.013
Unadjusted OR (95% CI)b
5.00 (1.51e16.51)
Adjusted OR (95% CI)c
4.91 (1.21e19.95)
CEA, carotid endarterectomy; CI, confidence interval; OR, odds ratio. a P value for between-group difference calculated using the Fisher’s exact or chi-squared tests. P < 0.05 is considered statistically significant. b Crude OR (95% CI) calculated using univariate logistic regression (i.e., considering the effect of each characteristic separately). c OR simultaneously adjusted for the factors that were found to be statistically significant in univariate analyses (i.e., the current adjusted, multivariate analysis considers the effect of female sex, the presence or absence of renal disease, and type of anesthesia simultaneously).
Table IV. Day and reason for hospital readmission of 8 patients in the 30-day postoperative period after 62 carotid endarterectomy procedures Reason
Other vascular surgery Endoprosthesis for abdominal aneurysm Contralateral carotid endarterectomy Femoral popliteal bypass procedure Open abdominal aneurysm repair Trauma Shoulder dislocation Severe hand trauma Medical Pneumonia Gastrointestinal bleeding and acenocoumarol intoxication
Postoperative day
7 30 30 30 21 21 14 30
intraoperative medical treatment,7,8 and finally to the extensive preoperative counseling of patients and their family with regard to early postoperative discharge. The latter practice is highly demanding because intentional early discharge after this type of delicate surgery requires a full explanation of
the perioperative risks, a high level of confidence between the physician and the patient and his/her family, and the availability of the patient’s relatives or friends in spite of potential interference with the activities of the latter. In addition, an efficient early discharge policy relies on full cooperation between the surgeon and the ward nursing staff to abide by the early discharge protocol to prevent undue delay of discharge. Kaufman et al.19 have previously investigated the feasibility and safety of discharge on the first postoperative day after CEA in a cohort of asymptomatic and symptomatic patients comparable to our study population. These authors achieved this goal in 60% of their patients. However, 13% of the patients in that study experienced complications within 30 days after CEA, including death in 3 patients and neurologic deficits in 5 patients. The poorer outcomes documents in this study, conducted >15 years ago, compared to the study presented here might be explained by an inferior medical therapy at that time. Nevertheless, Kaufman et al.19 also concluded that a short postoperative length of stay was not associated with a significant risk of unplanned readmission. According to the data of our prospective study, factors predisposing to postponed discharge were female sex, chronic renal insufficiency, and general
906 Van den Brande et al.
anesthesia. Postponed discharge of women was not attributable to medical reasons, nor to cardiovascular risk factors or comorbidities in any of the cases, but to the requirement of the study protocol to have someone available to stay with the patient after discharge; elderly patients were more frequently widowed and living alone than male patients. Collier20 reported that a clinical pathway including extensive preoperative patient education and a streamlined postoperative patient care permitted safe discharge on the first postoperative day in 84% of patients who had undergone CEA. This percentage is higher than that achieved in our study, but Collier’s study protocol20 did not require a relative to stay with the patient after discharge. It is possible that by omitting this social criterion the number of patients who can be discharged on the first postoperative day may increase. The identification of chronic renal insufficiency as a significant risk factor for a longer postoperative length of stay in our study has previously been reported.13,21 General anesthesia was the third factor associated with a prolonged postoperative stay in our study. Earlier studies have shown a reduced length of stay after CEA performed under local or regional anaesthesia.22e24 The superior postoperative neurocognitive outcome after CEA performed under local anesthesia may facilitate a quicker recovery and therefore earlier discharge.25 Other risk factors, including comorbidities and postoperative complications, such as preoperative neurologic symptoms, older age, diabetes mellitus, requirement of intravenous vasodilators, or postoperative neurologic or surgical morbidity14,15,21,26,27 that may preclude safe and early discharge after CEA were not significantly related to hospital length of stay in our study. None of the events of readmission to the hospital as recorded during the 30-day postoperative period were attributable to early discharge, which is in line with previous studies.19,21,28 Our present study concerns only a limited number of patients undergoing CEA, but is prospective and confirms most findings of former, mostly retrospective studies. A historical comparison in our department of length of postoperative stay before and during the present study was not attempted because before the study, patients were discharged at the discretion of the operating or ward surgeon, based only on clinical arguments. What is new here is that we looked for social issues that might render a safe and early discharge as inappropriate. Now, after the present study, we preoperatively propose this first postoperative day discharge to the patient if medically possible. However, we strongly advise older patients to have a relative to be with
Annals of Vascular Surgery
them for 24 hours. It was not the purpose of the study (or our intent) to calculate or to have impact on reimbursements or hospital resources. The goal was to increase patient satisfaction by early discharge but to be as safe as possible while doing so. Of special concern must be cerebral hyperperfusion syndrome. It is a relatively rare, but potentially devastating event that can occur a brief time after CEA. It is difficult to predict, probably because of increased arterial blood pressure, and occurs mostly in cases of nonelective CEA. Although the definite diagnosis has to be confirmed by imaging techniques, the presence of clinical symptoms is essential: severe headache, ocular or facial pain, confusion, visual disturbances, epileptic seizures, or any focal deficits not caused by cerebral ischemia. Awareness for these clinical signs and control of blood pressure are both part of postoperative patient surveillance. When present or when blood pressure is poorly controlled, early discharge is of course precluded.30,31 In the continuous evolution toward shortening of the perioperative length of stay for CEA, 1-day admission20 and same-evening discharge13,29 policies have also been adopted and have been shown to be feasible, but we believe that these policies may surpass the limits of safe surgical practice. CEA remains a controversial vascular surgical procedure, and any additional approach that could potentially reduce its efficacy should be viewed with extreme caution.32
CONCLUSION Discharge on the first postoperative day after elective CEA can be a safe and feasible policy in many patients, and unplanned readmissions linked to this early discharge must be uncommon. The social environment of the patient appeared to be the major factor preventing early discharge in our study. Appropriate preoperative patient and family counseling is essential before implementing an early discharge policy. Finally, in the continuous evolution of perioperative care of patients undergoing CEA, preoperative diagnostic work-up and risk factor management, perioperative antiplatelet therapy, experienced surgical and anesthetic practice, and excellent postoperative care must and will remain a real challenge in the additional reduction of perioperative neurologic morbidity and mortality. REFERENCES 1. North American Symptomatic Carotid Endarterectomy Trial Collaborators. Beneficial effect of carotid endarterectomy in symptomatic patients with high grade stenosis. N Engl Med 1991;325:445e8.
Vol. 28, No. 4, May 2014
2. Executive Committee for the Asymptomatic Carotid Atherosclerosis Study (ACAS). Endarterectomy for asymptomatic carotid artery stenosis. JAMA 1999;273:1424e8. 3. European Carotid Surgery Trialists’ Collaborative Group. Randomised trial of endarterectomy for recently symptomatic carotid stenosis: final results of the MRC European Carotid Surgery Trial (ECST). Lancet 1998;351:1379e87. 4. Berczi V, Randall M, Balamurugan R, et al. Safety of arch aortography for assessment of carotid arteries. Eur J Vasc Endovasc Surg 2006;31:3e7. 5. Borisch I, Horn M, Butz B, et al. Preoperative evaluation of carotid artery stenosis: comparison of contrast-enhanced MR angiography and duplex sonography with digital subtraction angiography. Am J Neuroradiol 2003;24:1117e22. 6. Johnston DC, Eastwood JD, Nguyen T, et al. Contrastenhanced magnetic resonance angiography of carotid arteries. Utility in routine clinical practice. Stroke 2002;33:2834e8. 7. Naylor AR. The importance of initiating ‘‘best medical therapy’’ and intervening as soon as possible in patients with symptomatic carotid artery disease: time for a radical rethink of practice. J Cardiovasc Surg 2009;50:773e82. 8. McGirt MJ, Perler BA, Brooke BS, et al. 3-hydroxy-3methylglutaryl coenzyme A reductase inhibitors reduce the risk of perioperative stroke and mortality after carotid endarterectomy. J Vasc Surg 2005;42:829e36. 9. Peeters W, Hellings WE, de Kleijn DP, et al. Carotid atherosclerotic plaques stabilize after stroke: insights into the natural process of atherosclerotic plaque stabilization. Arterioscler Thromb Vasc Biol 2009;29:128e33. 10. Liapis CD, Bell PRF, Mikhailidis D, et al. ESVS guidelines. Invasive treatment for carotid stenosis: indications and techniques. Eur J Vasc Endovasc Surg 2009;37(suppl 1):S1e19. 11. Hobart DC, Nicholas GG, Reed JF 3rd, et al. Carotid endarterectomy outcomes research: reduced resource utilization using a clinical protocol. Cardiovasc Surg 2000;8:446e51. 12. Angevine PD, Choudhri TF, Huang J, et al. Significant reductions in length of stay after carotid endarterectomy can be safely accomplished without modifying either anesthetic technique or postoperative ICU monitoring. Stroke 1999;30:2341e6. 13. Sheehan MK, Baker WH, Littooy FN, et al. Timing of postcarotid complications: a guide to safe discharge planning. J Vasc Surg 2001;34:13e6. 14. Roddy S, Estes J, Kwoun M, et al. Factors predicting prolonged length of stay after carotid endarterectomy. J Vasc Surg 2000;32:550e4. 15. Hernandez N, Salles-Cunha SX, Daoud YA, et al. Factors related to short length of stay after carotid endarterectomy. Vasc Endovasc Surg 2002;36:425e37. 16. Debing E, Aerden D, Van den Brande P. Diabetes mellitus is a predictor for early adverse outcome after carotid endarterectomy. Vasc Endovasc Surg 2011;45:28e32.
Discharge on postoperative day 1 after CEA 907
17. Debing E, Van den Brande P. Chronic renal insufficiency and risk of early mortality in patients undergoing carotid endarterectomy. Ann Vasc Surg 2006;20:609e13. 18. Debing E, Van den Brande P. Carotid endarterectomy in the elderly: are the patient characteristics, the early outcome, and the predictors the same as those in younger patients. Surg Neurol 2007;67:467e71. 19. Kaufman JL, Frank D, Rhee SW, et al. Feasibility and safety of 1-day postoperative hospitalization for carotid endarterectomy. Ann Surg 1996;131:751e5. 20. Collier PE. Are one-day admissions for carotid endarterectomy feasible? Am J Surg 1995;170:140e3. 21. Cikrit DF, Larson DM, Sawchuk AP, et al. Short- stay carotid endarterectomy in a tertiary-care Veterans Administration hospital. Am J Surg 2004;188:544e8. 22. McCarthy RJ, Walker R, McAteer P, et al. Patient and hospital benefits of local anaesthesia for carotid endarterectomy. Eur J Vasc Endovasc Surg 2001;22:13e8. 23. Mofidi R, Nimmo AF, Moores C, et al. Regional versus general anaesthesia for carotid endarterectomy: impact of change in practice. Surgeon 2006;4:158e62. 24. Kalko Y, Kafali E, Aydin U, et al. Surgery of the carotid artery: local anaesthesia versus general anaesthesia. Acta Chir Belg 2007;107:53e7. 25. Weber CF, Friedl H, Hueppe M, et al. Impact of general versus local anaesthesia on early postoperative cognitive dysfunction following carotid endarterectomy: GALA Study Subgroup Analysis. World J Surg 2009;33: 1526e32. 26. Katz SG, Kohl RD. Carotid endarterectomy with shortened hospital stay. Arch Surg 1995;130:887e90. 27. Young KC, Jahromi BS, Singh MJ, et al. Hospital resource use following carotid endarterectomy in 2006: analysis of the nationwide inpatient sample. J Stroke Cerebrovasc Dis 2010;19:458e64. 28. Su LT, Carpenter JP. Decreasing carotid endarterectomy length of stay at a university hospital. Cardiovasc Surg 1999;7:292e7. 29. Sheehan MK, Greisler HP, Littooy FN, et al. Same-evening discharge after carotid endarterectomy: our initial experience. J Vasc Surg 2004;39:575e7. 30. De Rango P. Cerebral hyperperfusion syndrome: the dark side of carotid endarterectomy. Eur J Vasc Endovasc Surg 2012;43:377. 31. Maas MB, Kwolek CJ, Hirsch JA, et al. Clinical risk predictors for cerebral hyperperfusion syndrome after carotid endarterectomy. J Neurol Neurosurg Psychiatry 2013;84: 569e72. 32. Ricotta JJ. Regarding: ‘‘Timing of postcarotid complications: a guide to safe discharge planning’’. J Vasc Surg 2001;34: 178e9.