- Email: [email protected]
Anatomic findings and outcomes associated with upper extremity arteriography and selective thrombolysis for acute finger ischemia
Anatomic findings and outcomes associated with upper extremity arteriography and selective thrombolysis for acute finger ischemia Arsalla Islam, MD,a Colston Edgerton, BS,a Jeanette M. Stafford, MS,b Andrew Koman, MD,c Zhongyu Li, MD,c Beth P. Smith, PhD,c Matthew S. Edwards, MD, MS,a and Matthew A. Corriere, MD, MS,a Winston-Salem, NC Objective: Limited evidence exists to guide clinical management of acute finger ischemia (AFI). To further inform diagnostic evaluation and decision making, we evaluated anatomic findings, procedural management, and amputation-free survival in an institutional cohort of patients with AFI. Methods: Consecutive patients undergoing transfemoral upper extremity angiography for AFI were identified. Clinical, laboratory, and procedural data were collected retrospectively from medical records, and arteriograms were reviewed to characterize anatomic findings. Telephone interviews were used to determine long-term outcomes, and additional symptomatic assessments (Symptom Severity and Functional Status scale, the Cold Sensitivity Severity scale, and the McGill Pain Severity Scale) were available in a subgroup of patients. Outcomes included anatomic findings, use of thrombolysis, complications, and amputation-free survival. Descriptive statistics and survival analysis were used to evaluate results. Results: Thirty-five patients (54% women) were analyzed with a median follow-up of 13.7 months. Symptom duration at time of presentation ranged from 1 to 28 days, and seven patients had tissue loss or gangrene, or both. Mean age was 47.7 6 12.2 years. Baseline characteristics included smoking in 22 (65%), connective tissue disorder in 11 (31%), and history of repetitive hand trauma in 10 (29%). The most frequent anatomic location of arterial pathology identified during angiography was distal to the wrist (n [ 32), including eight ulnar/radial aneurysms; upper arm (n [ 3) and forearm (n [ 8) lesions were less common. Sixteen patients were treated with catheter-directed thrombolysis, of which eight (50%) had interval anatomic improvement on repeat angiography. Procedure-related adverse events associated with angiography included bleeding (n [ 3) and pseudoaneurysm (n [ 1). Eleven of 35 patients had subsequent surgical revascularization at a median of 15 days after angiography. Estimated (standard error) amputation-free survival was 0.88 (0.07) at 1 month and 0.84 (0.08) at 6 months among patients without tissue loss or gangrene. Estimated 60-day amputation-free survival was 0.84 (standard error, 0.08). Overall amputation-free survival was similar between patients managed with vs without thrombolysis (P [ .61), but subgroup analysis of those patients without tissue loss or gangrene at the time of presentation revealed a trend toward improved amputation-free survival with use of thrombolysis, with 60-day amputation-free survival of 0.92 vs 0.75 (P [ .12). Persistent late symptoms were present in 17 patients (48.6%) at the last follow-up and were generally characterized as mild by functional and pain scale assessments. Conclusions: Angiography performed for AFI frequently identifies distal occlusive disease, and catheter-directed thrombolysis may expand revascularization options in select patients. (J Vasc Surg 2014;-:1-8.)
Acute finger ischemia (AFI) is an uncommon problem associated with significant morbidity. Anatomic localization of the occlusive disease is necessary to guide management of AFI, including decisions for revascularization vs debridement or digital amputation. Surgical embolectomy has been widely practiced for the treatment of upper From the Departments of Vascular and Endovascular Surgery,a Biostatistical Sciences,b and Orthopedic Surgery,c Wake Forest University School of Medicine. Author conflict of interest: none. Reprint requests: Matthew A. Corriere, MD, MS, Wake Forest University Baptist Medical Center, Department of Vascular and Endovascular Surgery, Medical Center Blvd, Winston-Salem, NC 27157 (e-mail: [email protected]). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest. 0741-5214/$36.00 Copyright Ó 2014 by the Society for Vascular Surgery. http://dx.doi.org/10.1016/j.jvs.2014.02.005
extremity thromboembolism.1 More recently, the use of thrombolysis (including catheter-directed thrombolysis)2-7 has also been described, but clinical outcomes associated with either management approach for AFI have not been well characterized. Because the anatomic disease distribution, natural history, and treatment outcomes associated with AFI have not been thoroughly described, limited evidence exists to guide diagnostic evaluation and treatment of patients presenting with this diagnosis. To further characterize outcomes and inform clinical decision making, we retrospectively evaluated a cohort of consecutive patients undergoing upper extremity arteriography for AFI at a single center during a 5-year period. METHODS This retrospective single-center study was performed with approval from the Wake Forest University School of Medicine Institutional Review Board. 1
JOURNAL OF VASCULAR SURGERY --- 2014
2 Islam et al
Fig 1. Algorithm shows the workup of patients with acute finger ischemia (AFI) at our institution.
Patient cohort identification and data collection. Consecutive patients with AFI evaluated with transfemoral upper extremity arteriography were identified from an operative registry. An algorithm for the workup of patients with AFI ischemia at our institution is provided in Fig 1. AFI was diagnosed by the vascular surgeon performing angiography according to symptoms and clinical examination. An interval of #30 days between symptom onset and presentation was used to define AFI, and patients undergoing arteriography in the setting of longer symptom duration were excluded. Other exclusion criteria included AFI in the setting of a previous ipsilateral upper extremity arteriovenous hemodialysis access procedure, ischemia secondary to trauma or iatrogenic injury, or global upper extremity ischemia in the setting of suspected brachial artery thromboembolism, which is routinely managed at our institution by brachial artery exploration and thrombectomy without angiography.
Data collection included review of medical records (including notes, laboratory results, and demographic data) and electronically archived arteriograms. Arteriograms and operative notes were assessed for the following specific end points: type and anatomic level of arterial lesion(s) identified, use of intra-arterial thrombolysis, and performance of any endovascular interventions (such as balloon angioplasty or stenting, or both) during the same procedure. Among patients treated with transcatheter intra-arterial thrombolytic infusions, repeat arteriograms were designated as improved if vessels that were occluded at the initial angiography were patent on repeat imaging. Patients were also contacted and interviewed by telephone to determine symptomatic status and identify any subsequent interventions not captured within our institutional records. Additional symptomatic assessments using standardized instruments, including the Symptom Severity and
JOURNAL OF VASCULAR SURGERY Volume -, Number -
Islam et al 3
Fig 2. Anatomic improvement is seen with catheter-directed thrombolysis. a, Arteriogram at presentation demonstrates thrombosed ulnar artery. b, Repeat arteriogram after 24-hour infusion of catheter-directed thrombolysis demonstrates restored ulnar artery patency with evidence of an ulnar artery aneurysm. The patient underwent subsequent surgical repair of the ulnar aneurysm after a brief period of anticoagulation.
Functional Status scale, the Cold Sensitivity Severity scale, and the McGill Pain Severity Scale were available for a subset of patients. Validity of these instruments for hand symptom assessment and correlation with health-related quality of life have been described previously.8-10 Procedural management. Catheter angiography was used as the primary imaging modality after clinical diagnosis of AFI, and all angiography procedures were performed by a group of five surgeons. All angiography procedures were performed through ultrasound-guided common femoral artery access established using a micropuncture technique before upsizing to a short 5F sheath. Before aortic arch instrumentation, patients were systemically anticoagulated with an 80 U/kg intravenous bolus of unfractionated heparin and redosed as needed to maintain an activated clotting time >280 seconds. Aortic arch angiography was then performed, followed by selective access of the ipsilateral subclavian artery using a 0.035-inch wire and a diagnostic angiography catheter. Selective arteriograms were performed using handheld injections of dilute contrast with digital subtraction. Intra-arterial nitroglycerine (200-400 mg) was given at the discretion of the operator if significant catheterinduced vasospasm occurred. Intra-arterial thrombolysis was performed at the discretion of the operating surgeon and was typically used when angiography identified one of the following scenarios: evidence of radial or ulnar artery aneurysm with distal embolization (Fig 2), pruning of the proper digital vessels consistent with small-vessel thrombosis, or lack of surgical revascularization alternatives for finger salvage due to absence of arterial outflow. Thrombolytic infusions were typically avoided in the setting of small-diameter brachial arteries, evidence of procedure-induced brachial artery spasm, or when primary endovascular or surgical treatment was deemed most appropriate.
Intra-arterial pharmacologic thrombolysis was performed using alteplase administered as an initial 6- to 10-mg bolus, followed by a continuous infusion at 1 mg/h through an indwelling arterial catheter placed in the axillary or brachial artery. All patients treated with thrombolytic infusions were monitored in an intensive care unit and studied with repeat angiography performed through the infusion catheter within 12 to 24 hours. Further thrombolysis was performed selectively based on a combination of anatomic findings, symptomatic response, clinical examination, and laboratory data. Sheath removal was performed after confirmation of an activated clotting time <180 seconds, with manual compression applied for 20 minutes to establish hemostasis. In patients treated with thrombolysis, sheath removal occurred at a minimum of 4 hours from cessation of thrombolytic infusion. Patients with suspected vasospasm or limited options for revascularization, or both, were routinely treated with amlodipine (5-10 mg/d). Decisions regarding subsequent anticoagulation, surgical revascularization, or digital amputation were made based on clinical examination and arteriogram findings by a multidisciplinary team of hand and vascular surgery specialists. Evaluations of symptom severity, functional status, and cold sensitivity severity8-10 were available in a subgroup of the patients treated with separate surgical procedures (revascularization or periarterial sympathectomy, or both) after angiography or thrombolysis, or both. These validated scales have demonstrated reliability for quantifying upper extremity symptoms and health-related quality of life (HRQL).8 Statistical analysis. Descriptive statistics are reported as counts (%) for categoric variables and mean 6 standard deviation or median (interquartile range) for continuous variables. Kaplan-Meier survival analysis was used to
JOURNAL OF VASCULAR SURGERY --- 2014
4 Islam et al
Table I. Patient demographics and clinical characteristics Characteristic
No. (%) or mean 6 SD
Age, years 47.7 6 12.2 Female sex 19 (54.3) Nonwhite race/ethnicity 4 (11.4) a 14 (41.2) Current smoker a 22 (64.7) Ever smoker Connective tissue disorder 11 (31.4) Occupation-related repetitive hand or wrist traumab 10 (28.6) Hypercoagulable 5 (14.3) Diabetes 4 (11.4) Atrial fibrillation 1 (2.9) Intravenous drug abuse 1 (2.9) Tissue loss/gangrene 7 (20.0) Symptom laterality Right 17 (48.6) Left 16 (45.7) Bilateral 2 (5.7) SD, Standard deviation. a Smoking status unavailable for one patient. b Occupations involving repetitive hand or wrist trauma included mechanics, transcriptionists, and construction workers.
evaluate amputation-free survival, and comparisons based on use of thrombolysis were performed using the log-rank test. Survival estimates are presented as 60-day estimate with standard error. All statistical analyses were performed using SAS 9.3 software (SAS Institute, Cary, NC). RESULTS The study identified and included 35 patients (54% female) who underwent arteriography for AFI during the 5-year period from 2008 to 2013. Overall median follow up was 13.7 months (interquartile range, 3.619.2 months). Mean patient age at initial presentation was 47.7 6 12.2 years. History of current or previous smoking was highly prevalent (41% and 65%, respectively), as were a history of connective tissue disorder (31%), chronic history of occupation-related repetitive hand trauma (29%), and hypercoagulable state (14%; Table I). The presentation in one patient was consistent thromboangiitis obliterans (Buerger disease). Patients presented with nearly equal frequencies of right-handed vs left-handed symptoms, and seven patients had gangrene or ulceration at the initial presentation. Catheter angiography was the initial anatomic imaging study in 24 patients (69%), and 11 patients (31%) had been evaluated with computed tomography angiography before catheter angiography. Angiography findings are summarized in Table II. Anatomic variants included incomplete palmar arch in 22 patients, proximal (upper arm) brachial artery bifurcation in 7, and radial or ulnar artery aneurysms, or both, in 8. Occlusive lesions were identified distal to the wrist joint in 32 patients (94.1%), whereas lesions were less frequently identified at the level of the forearm (n ¼ 8), upper arm (n ¼ 3), and subclavian artery (n ¼ 2). Among the 28 patients with unilateral AFI who were studied with bilateral
Table II. Arteriogram findings Finding Level of occlusive disease Hand only Forearm and hand Subclavian and hand Proximal arm, forearm, and hand Proximal arm and forearm Proximal arm only Subclavian only Incomplete palmar archa Contralateral upper extremity diseaseb Ulnar or radial aneurysm Upper arm brachial artery bifurcation
Frequency (%) 24 6 1 1 1 1 1 22 9 8 7
(68.6) (17.1) (2.9) (2.9) (2.9) (2.9) (2.9) (64.7) (32.1) (22.9) (20.0)
a
The palmar arch in one patient was not visualized on angiography. Among 28 patients with unilateral symptoms who were evaluated with bilateral upper extremity angiography. b
upper extremity angiography, nine (32.1%) had bilateral disease. Five patients underwent procedural interventions at the initial angiography, including thromboembolectomy (n ¼ 3) and subclavian artery angioplasty and stenting (n ¼ 2). Sixteen patients were treated with catheter-directed thrombolysis and studied with repeat angiography, which revealed interval anatomic improvement in eight (50%; Fig 2). Diagnostic angiography was performed without simultaneous procedural intervention or catheter-directed thrombolysis in 13 patients (37.1%), none of whom had gangrene or tissue loss at the time of presentation. Perioperative complications after angiography occurred in four patients (11.4%), including 2 access site hematomas, 1 femoral pseudoaneurysm, and 1 hematoma at the site of a peripheral intravenous catheter. The four patients with complications had been treated with catheter-directed thrombolytic infusions, and the infusions in three were discontinued after 8 to 24 hours when the complication was recognized. None of the patients with angiographyrelated complications required a blood transfusion or surgical exploration for bleeding. The femoral pseudoaneurysm occurred in a patient who was therapeutically anticoagulated with warfarin while awaiting surgical revascularization and was treated electively with ultrasound-guided thrombin injection 2 weeks after angiography. Eleven patients underwent subsequent surgical revascularization at a median of 15 days after angiography (Table III). These procedures included radial or ulnar aneurysm repair (n ¼ 7), radial or ulnar artery bypass (n ¼ 2), ulnar aneurysm excision (n ¼ 1), and radial artery thrombectomy (n ¼ 1). Adjunctive periarterial sympathectomy was performed in addition to revascularization in nine of 11 patients, and five additional patients who were not candidates for revascularization underwent periarterial sympathectomy alone. Eight patients underwent finger amputation during the follow-up period, seven of which were performed #60 days of initial presentation with tissue loss. Three of the eight finger amputations (37.5%) were performed on patients
JOURNAL OF VASCULAR SURGERY Volume -, Number -
Islam et al 5
Table III. Subsequent procedures stratified by use of catheter-directed thrombolytic infusion after initial angiographya Thrombolytic infusion Subsequent surgical procedure None Surgical revascularization þ sympathectomy Finger amputation only Sympathectomy only Surgical revascularization only Sympathectomy þ finger amputation
Yes (n ¼ 16)
No (n ¼ 19)
5 (31.3) 7 (43.8)
10 (52.6) 15 (42.9) 2 (10.5) 9 (25.7)
1 0 2 1
(6.3) (0) (12.5) (6.3)
3 3 0 1
(15.8) (15.8) (0) (5.3)
Total (N ¼ 35)
4 3 2 2
(11.4) (8.6) (5.7) (5.7)
a All procedures were performed in a delayed fashion in a separate procedural setting from angiography. Data displayed as frequency (%) by category.
whose initial arteriogram was not accompanied by simultaneous thrombectomy, endovascular intervention, or thrombolysis. Estimated 60-day amputation-free survival was 0.78 (standard error, 0.07; Fig 3). Overall amputation-free survival was similar between patients managed with vs without thrombolysis (P ¼ .61; Fig 4, A). Subgroup analysis of those patients without tissue loss or gangrene at presentation revealed a trend toward improved amputation-free survival with use of thrombolysis, with 60-day amputation-free survival of 0.92 vs 0.75 (P ¼ .12; Fig 4, B). No patients treated with surgical revascularization underwent subsequent finger amputation. At the time of last follow-up, 17 patients (48.6%) reported that they were symptom-free, 16 (45.7%) reported ongoing symptoms, and symptomatic status could not be verified in two patients. Contralateral hand symptoms developed during late follow-up in seven patients, five of whom had been evaluated with bilateral angiography and two of whom had been evaluated with unilateral angiography at the initial presentation. Symptomatic, functional, and pain scale assessments were performed at a median of 76 days, and cold sensitivity scores were performed at median of 142 days after presentation. These evaluations are summarized in Table IV and generally revealed persistent but mild symptoms. DISCUSSION AFI is a relatively uncommon condition that requires consideration of a wide variety of diagnoses, including atherosclerosis, embolization (from proximal and distal sources), vasculitis, vasospasm, thromboangiitis obliterans (Buerger disease), and thoracic outlet syndrome. Limited evidence exists to guide decision making related to evaluation and management of AFI, and the 35 patients described in this analysis represent, to our knowledge, the largest North American series reported to date. Although our analysis is limited by a retrospective design and relatively small patient cohort, we believe that the
Fig 3. Estimated overall amputation-free survival. The solid line indicates the proportion of amputation-free patients, and the vertical marks indicate censoring at time of last follow up. The number of patients at risk is indicated along the x-axis.
results provide several observations that potentially inform diagnostic evaluation and clinical management. Consistent with other previous reports, our cohort had approximately equal numbers of men and women with a high prevalence of smoking, connective tissue disorders, thrombophilia, and occupation-related repetitive hand trauma. In contrast to other reports describing a high incidence of cardioembolic AFI,11 however, only one of 35 patients in our series had a history of atrial fibrillation. Occlusive disease of the innominate or subclavian arteries, or both, was also relatively uncommon, whereas lesions distal to the wrist were identified in 32 of 35 patients. We believe that the predominately distal anatomic distribution of the pathology identified in this series supports use of angiography as the primary imaging modality for evaluation of AFI, primarily due to advantages over cross-sectional imaging for characterizing occlusive disease of the palmar or digital vessels, or both. In addition, angiography permits endovascular treatment during the same procedure when anatomically suitable disease is identified. Bilateral upper extremity arterial occlusive disease was identified in approximately one-third of patients with unilateral AFI who were studied with bilateral angiography, suggesting that AFI often occurs in the setting of chronic systemic disease that is not limited to the symptomatic extremity. Given that bilateral symptoms developed during follow-up in more than half of the patients with asymptomatic contralateral disease identified by angiography, our data also suggest that identification of contralateral disease warrants close surveillance with aggressive medical management. Additional studies are needed to determine whether a pre-emptive procedural intervention affects long-term
6 Islam et al
JOURNAL OF VASCULAR SURGERY --- 2014
Fig 4. Estimated amputation-free survival stratified by use of catheter-directed thrombolysis. A, Overall cohort (P ¼ .61). B, Patient subgroup without tissue loss or gangrene at the time of presentation (P ¼ .12).
risk of symptoms or finger amputation in the setting of asymptomatic disease. One-half of the patients treated with thrombolytic infusion had interval anatomic improvement when evaluated with repeat angiography #24 hours. Although this rate of anatomic improvement suggests that thrombolysis may improve amputation-free survival or expand anatomic options for subsequent revascularization procedures, or both, it is also important to note that thrombolysis was used in all four patients who experienced perioperative complications after angiography. Although these complications were access-site related and no patients required transfusion, these results underscore the importance of appropriate preoperative patient counseling and suggest that thrombolysis should be used selectively based on anticipated risk and potential benefit. Our results suggest that thrombolysis may be most beneficial among patients without tissue loss or digital gangrene; presumably, thrombolysis in these patients may reestablish patency of small-caliber distal arteries, which are critical for finger salvage. Accordingly, we believe that thrombolysis may be particularly useful in the setting of digital embolism, where it may reestablish patency of digital arteries poorly suited to direct surgical revascularization (such as bypass or thrombectomy, or both) due to their small diameter and potential for vasospasm with manipulation. The role of vascular reconstruction after thrombolytic therapy depends on the anatomic results and presumed etiology. Patients with radial or ulnar artery occlusion, or both, and adequate collateral circulation often improve with thrombolysis, presumably through elimination of on-going thromboembolism and reduction of associated vasospasm, before elective resection or reconstruction. In patients with inadequate collaterals and unreconstructable distal disease, thrombolysis has the potential to improve distal outflow and collateral flow, converting some patients to candidates for reconstruction and decreasing procedural complexity by reducing graft length. At our institution, reconstruction is not performed immediately to avoid the potential for residual vasospasm resulting from proximal
Table IV. Function, cold sensitivity, and pain scores (n ¼ 12)a Instrument (range of possible scores) McCabe cold sensitivity severity scale score (0-400) Levine 11-item symptom mean score (1-5) Levine 8-item function mean score (1-5) McGill pain score (visual scale scored 0-10)
Median (Q1, Q3) 254 (170, 325) 1.9 (1.5, 3.0) 1.6 (1.0, 2.4) 4.0 (0.5, 6.0)
Q1, Lower quartile; Q3, upper quartile. a Ranges for each score are displayed next to the instrument name. Symptomatic, functional, and pain scale assessments were performed at a median of 76 days, and cold sensitivity scores were performed at median of 142 days after following presentation.
arterial catheterization. In a previous report from our institution, Chloros et al12 observed a 2-year primary patency of 77% in a retrospective cohort of 13 hands in 12 patients treated with reversed interposition vein grafts for both acute and chronic disease.12 After angiography or thrombolysis, or both, for AFI in the current series, subsequent revascularization or periarterial sympathectomy, or both, were performed with greater frequency than digital amputation. Persistent long-term symptoms and functional impairment were identified in approximately half of the patients. At our center, periarterial sympathectomy is used for nonhealing digital ulcers when revascularization is not possible or as an adjunct to revascularization in patients with vasospastic disease. We prefer periarterial sympathectomy rather than central sympathectomy based on experience with this technique at our institution and the ability to combine this procedure with local revascularization or debridement, or both, under the same regional anesthetic. Patients who were not candidates for lytic treatment had no suitable targets identified, had small diameter vessels with pruning of the digital vessels, and were managed expectantly. Because pain and functional assessment scale assessments in this study were only obtained for patients
JOURNAL OF VASCULAR SURGERY Volume -, Number -
who underwent surgical revascularization or sympathectomy, or both, however, we are unable to determine how these outcomes differ in patients managed with endovascular procedures, primary amputation, or medical therapy alone. Treatment of finger ischemia with thrombolysis has been described.2-4,13 Coulon et al2 reported a series of 13 patients treated with streptokinase, and eight of 13 patients in that series had complete clot dissolution, whereas partial clearance was seen in the remainder. A high incidence of complications was also observed in their series, including six access-site hematomas, one extremity hematoma attributed to catheter-related trauma, and one transient ischemic attack. Cejna et al4 reported a retrospective analysis of 38 patients undergoing thrombolysis for acute upper extremity ischemia. These authors observed a similar anatomic disease distribution as described in this report, including 3 isolated upper arm lesions, 9 patients with combined brachial and forearm lesions, and 28 lesions located in the forearm or hand arteries, or both. Complete recanalization was achieved in proximal subclavian, axillary, and isolated brachial arterial occlusions, resulting in a success rate of 100%; however, the rate of angiographic success (defined as complete resolution of flow to the hand) was only 33%. These authors defined successful thrombolysis based on restoration of blood flow to at least one patent forearm artery supplying the palmar arch and concluded that thrombolysis results were comparable to surgical embolectomy. Miyayama et al3 described the results of catheterdirected aspiration in four patients with atrial fibrillation and upper extremity embolism, of whom three were also treated with thrombolysis. No major extremity amputations were required, and one procedure-related arterial perforation was managed with coil embolization. Several additional limitations of our analysis deserve specific mention. First, our use of a symptom duration of #30 days to define acute ischemia might be considered overly inclusive. We chose this cut point based on the general impression that, in contrast to acute lower extremity ischemia or more global upper extremity ischemia (as may result from a brachial artery embolism), AFI presentations are frequently delayed, particularly in the absence of a major neurologic deficit or gangrene. Use of a shorter interval after symptom onset might have more precisely defined the acuity of symptoms but also would have potentially excluded some patients more appropriately categorized as acute rather than chronic. Unfortunately, due to the rare nature of AFI, there is no widely accepted definition of acuity-based duration of symptoms. Second, our study excluded patients with AFI in the setting of a previous upper extremity arteriovenous hemodialysis access or major extremity trauma because distinct diagnostic considerations, management strategies, and associated outcomes for these scenarios have been described. These exclusion criteria therefore limit generalization of our results and do not inform management of patients with these diagnoses.
Islam et al 7
Third, postintervention surveillance likely resulted in longer and more detailed follow-up among patients treated with surgical or endovascular interventions, potentially biasing our results toward missed late events in patients who did not undergo revascularization procedures. Finally, we must acknowledge that patients evaluated at our hospital for AFI in departments other than ours (including those studied without catheter angiography) would have been omitted from this analysis due to our use of catheter angiography to define the study cohort. This potential selection bias might have influenced our outcomes by excluding patients with less severe symptoms or occlusive disease, or both, at the time of initial presentation. CONCLUSIONS Despite these limitations, however, this analysis provides important data related to presentation and procedural outcomes associated with AFI. Our results potentially inform decision making related to diagnostic imaging and clinical management of this uncommon problem. AUTHOR CONTRIBUTIONS Conception and design: MC, AI, ME Analysis and interpretation: AI, MC, ME Data collection: AI, CE Writing the article: AI, MC, JS Critical revision of the article: MC, AI, AK, ZL, BS Final approval of the article: MC, AI, JS, ME Statistical analysis: JS Obtained funding: Not applicable Overall responsibility: MC REFERENCES 1. Andersen LV, Mortensen LS, Lindholt JS, Faergeman O, Henneberg EW, Frost L. Upper-limb thrombo-embolectomy: national cohort study in Denmark. Eur J Vasc Endovasc Surg 2010;40:628-34. 2. Coulon M, Goffette P, Dondelinger RF. Local thrombolytic infusion in arterial ischemia of the upper limb: mid-term results. Cardiovasc Intervent Radiol 1994;17:81-6. 3. Miyayama S, Yamashiro M, Shibata Y, Hashimoto M, Yoshida M, Tsuji K, et al. Thrombolysis and thromboaspiration for acute thromboembolic occlusion in the upper extremity. Jpn J Radiol 2012;30: 180-4. 4. Cejna M, Salomonowitz E, Wohlschlager H, Zwrtek K, Bock R, Zwrtek R. rt-PA thrombolysis in acute thromboembolic upperextremity arterial occlusion. Cardiovasc Intervent Radiol 2001;24: 218-23. 5. Barbiero G, Cognolato D, Casarin A, Guarise A. Intra-arterial thrombolysis of acute hand ischaemia with or without microcatheter: preliminary experience and comparison with the literature. Radiol Med 2011;116:919-31. 6. Friedrich KM, Fruhwald-Pallamar J, Stadlbauer A, Salem G, Salomonowitz E. Hypothenar hammer syndrome: long-term follow-up of selective thrombolysis by 3.0-T MR angiography. Eur J Radiol 2010;75:e27-31. 7. Hering J, Angelkort B. [Acute ischemia of the hand after intra-arterial injection of flunitrazepam. Local combined fibrinolysis therapy in three cases]. Dtsch Med Wochenschr 2006;131:1377-80. 8. Ruch DS, Smith TL, Smith BP, Holden M, Russell G, Koman LA. Anatomic and physiologic evaluation of upper extremity ischemia. Microsurgery 1999;19:181-8.
8 Islam et al
9. Levine DW, Simmons BP, Koris MJ, Daltroy LH, Hohl GG, Fossel AH, et al. A self-administered questionnaire for the assessment of severity of symptoms and functional status in carpal tunnel syndrome. J Bone Joint Surg Am 1993;75:1585-92. 10. McCabe SJ, Mizgala C, Glickman L. The measurement of cold sensitivity of the hand. J Hand Surg Am 1991;16:1037-40. 11. Andersen LV, Mortensen LS, Lip GY, Lindholt JS, Faergeman O, Henneberg EW, et al. Atrial fibrillation and upper limb thromboembolectomy: a national cohort study. J Thromb Haemost 2011;9:1738-43.
JOURNAL OF VASCULAR SURGERY --- 2014
12. Chloros GD, Lucas RM, Li Z, Holden MB, Koman LA. Post-traumatic ulnar artery thrombosis: outcome of arterial reconstruction using reverse interpositional vein grafting at 2 years minimum follow-up. J Hand Surg 2008;33:932-40. 13. Deguara J, Ali T, Modarai B, Burnand KG. Upper limb ischemia: 20 years experience from a single center. Vascular 2005;13:84-91.
Submitted Oct 22, 2013; accepted Feb 4, 2014.
Acute finger ischemia (AFI) is an uncommon problem associated with significant morbidity. Anatomic localization of the occlusive disease is necessary to guide management of AFI, including decisions for revascularization vs debridement or digital amputation. Surgical embolectomy has been widely practiced for the treatment of upper From the Departments of Vascular and Endovascular Surgery,a Biostatistical Sciences,b and Orthopedic Surgery,c Wake Forest University School of Medicine. Author conflict of interest: none. Reprint requests: Matthew A. Corriere, MD, MS, Wake Forest University Baptist Medical Center, Department of Vascular and Endovascular Surgery, Medical Center Blvd, Winston-Salem, NC 27157 (e-mail: [email protected]). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest. 0741-5214/$36.00 Copyright Ó 2014 by the Society for Vascular Surgery. http://dx.doi.org/10.1016/j.jvs.2014.02.005
extremity thromboembolism.1 More recently, the use of thrombolysis (including catheter-directed thrombolysis)2-7 has also been described, but clinical outcomes associated with either management approach for AFI have not been well characterized. Because the anatomic disease distribution, natural history, and treatment outcomes associated with AFI have not been thoroughly described, limited evidence exists to guide diagnostic evaluation and treatment of patients presenting with this diagnosis. To further characterize outcomes and inform clinical decision making, we retrospectively evaluated a cohort of consecutive patients undergoing upper extremity arteriography for AFI at a single center during a 5-year period. METHODS This retrospective single-center study was performed with approval from the Wake Forest University School of Medicine Institutional Review Board. 1
JOURNAL OF VASCULAR SURGERY --- 2014
2 Islam et al
Fig 1. Algorithm shows the workup of patients with acute finger ischemia (AFI) at our institution.
Patient cohort identification and data collection. Consecutive patients with AFI evaluated with transfemoral upper extremity arteriography were identified from an operative registry. An algorithm for the workup of patients with AFI ischemia at our institution is provided in Fig 1. AFI was diagnosed by the vascular surgeon performing angiography according to symptoms and clinical examination. An interval of #30 days between symptom onset and presentation was used to define AFI, and patients undergoing arteriography in the setting of longer symptom duration were excluded. Other exclusion criteria included AFI in the setting of a previous ipsilateral upper extremity arteriovenous hemodialysis access procedure, ischemia secondary to trauma or iatrogenic injury, or global upper extremity ischemia in the setting of suspected brachial artery thromboembolism, which is routinely managed at our institution by brachial artery exploration and thrombectomy without angiography.
Data collection included review of medical records (including notes, laboratory results, and demographic data) and electronically archived arteriograms. Arteriograms and operative notes were assessed for the following specific end points: type and anatomic level of arterial lesion(s) identified, use of intra-arterial thrombolysis, and performance of any endovascular interventions (such as balloon angioplasty or stenting, or both) during the same procedure. Among patients treated with transcatheter intra-arterial thrombolytic infusions, repeat arteriograms were designated as improved if vessels that were occluded at the initial angiography were patent on repeat imaging. Patients were also contacted and interviewed by telephone to determine symptomatic status and identify any subsequent interventions not captured within our institutional records. Additional symptomatic assessments using standardized instruments, including the Symptom Severity and
JOURNAL OF VASCULAR SURGERY Volume -, Number -
Islam et al 3
Fig 2. Anatomic improvement is seen with catheter-directed thrombolysis. a, Arteriogram at presentation demonstrates thrombosed ulnar artery. b, Repeat arteriogram after 24-hour infusion of catheter-directed thrombolysis demonstrates restored ulnar artery patency with evidence of an ulnar artery aneurysm. The patient underwent subsequent surgical repair of the ulnar aneurysm after a brief period of anticoagulation.
Functional Status scale, the Cold Sensitivity Severity scale, and the McGill Pain Severity Scale were available for a subset of patients. Validity of these instruments for hand symptom assessment and correlation with health-related quality of life have been described previously.8-10 Procedural management. Catheter angiography was used as the primary imaging modality after clinical diagnosis of AFI, and all angiography procedures were performed by a group of five surgeons. All angiography procedures were performed through ultrasound-guided common femoral artery access established using a micropuncture technique before upsizing to a short 5F sheath. Before aortic arch instrumentation, patients were systemically anticoagulated with an 80 U/kg intravenous bolus of unfractionated heparin and redosed as needed to maintain an activated clotting time >280 seconds. Aortic arch angiography was then performed, followed by selective access of the ipsilateral subclavian artery using a 0.035-inch wire and a diagnostic angiography catheter. Selective arteriograms were performed using handheld injections of dilute contrast with digital subtraction. Intra-arterial nitroglycerine (200-400 mg) was given at the discretion of the operator if significant catheterinduced vasospasm occurred. Intra-arterial thrombolysis was performed at the discretion of the operating surgeon and was typically used when angiography identified one of the following scenarios: evidence of radial or ulnar artery aneurysm with distal embolization (Fig 2), pruning of the proper digital vessels consistent with small-vessel thrombosis, or lack of surgical revascularization alternatives for finger salvage due to absence of arterial outflow. Thrombolytic infusions were typically avoided in the setting of small-diameter brachial arteries, evidence of procedure-induced brachial artery spasm, or when primary endovascular or surgical treatment was deemed most appropriate.
Intra-arterial pharmacologic thrombolysis was performed using alteplase administered as an initial 6- to 10-mg bolus, followed by a continuous infusion at 1 mg/h through an indwelling arterial catheter placed in the axillary or brachial artery. All patients treated with thrombolytic infusions were monitored in an intensive care unit and studied with repeat angiography performed through the infusion catheter within 12 to 24 hours. Further thrombolysis was performed selectively based on a combination of anatomic findings, symptomatic response, clinical examination, and laboratory data. Sheath removal was performed after confirmation of an activated clotting time <180 seconds, with manual compression applied for 20 minutes to establish hemostasis. In patients treated with thrombolysis, sheath removal occurred at a minimum of 4 hours from cessation of thrombolytic infusion. Patients with suspected vasospasm or limited options for revascularization, or both, were routinely treated with amlodipine (5-10 mg/d). Decisions regarding subsequent anticoagulation, surgical revascularization, or digital amputation were made based on clinical examination and arteriogram findings by a multidisciplinary team of hand and vascular surgery specialists. Evaluations of symptom severity, functional status, and cold sensitivity severity8-10 were available in a subgroup of the patients treated with separate surgical procedures (revascularization or periarterial sympathectomy, or both) after angiography or thrombolysis, or both. These validated scales have demonstrated reliability for quantifying upper extremity symptoms and health-related quality of life (HRQL).8 Statistical analysis. Descriptive statistics are reported as counts (%) for categoric variables and mean 6 standard deviation or median (interquartile range) for continuous variables. Kaplan-Meier survival analysis was used to
JOURNAL OF VASCULAR SURGERY --- 2014
4 Islam et al
Table I. Patient demographics and clinical characteristics Characteristic
No. (%) or mean 6 SD
Age, years 47.7 6 12.2 Female sex 19 (54.3) Nonwhite race/ethnicity 4 (11.4) a 14 (41.2) Current smoker a 22 (64.7) Ever smoker Connective tissue disorder 11 (31.4) Occupation-related repetitive hand or wrist traumab 10 (28.6) Hypercoagulable 5 (14.3) Diabetes 4 (11.4) Atrial fibrillation 1 (2.9) Intravenous drug abuse 1 (2.9) Tissue loss/gangrene 7 (20.0) Symptom laterality Right 17 (48.6) Left 16 (45.7) Bilateral 2 (5.7) SD, Standard deviation. a Smoking status unavailable for one patient. b Occupations involving repetitive hand or wrist trauma included mechanics, transcriptionists, and construction workers.
evaluate amputation-free survival, and comparisons based on use of thrombolysis were performed using the log-rank test. Survival estimates are presented as 60-day estimate with standard error. All statistical analyses were performed using SAS 9.3 software (SAS Institute, Cary, NC). RESULTS The study identified and included 35 patients (54% female) who underwent arteriography for AFI during the 5-year period from 2008 to 2013. Overall median follow up was 13.7 months (interquartile range, 3.619.2 months). Mean patient age at initial presentation was 47.7 6 12.2 years. History of current or previous smoking was highly prevalent (41% and 65%, respectively), as were a history of connective tissue disorder (31%), chronic history of occupation-related repetitive hand trauma (29%), and hypercoagulable state (14%; Table I). The presentation in one patient was consistent thromboangiitis obliterans (Buerger disease). Patients presented with nearly equal frequencies of right-handed vs left-handed symptoms, and seven patients had gangrene or ulceration at the initial presentation. Catheter angiography was the initial anatomic imaging study in 24 patients (69%), and 11 patients (31%) had been evaluated with computed tomography angiography before catheter angiography. Angiography findings are summarized in Table II. Anatomic variants included incomplete palmar arch in 22 patients, proximal (upper arm) brachial artery bifurcation in 7, and radial or ulnar artery aneurysms, or both, in 8. Occlusive lesions were identified distal to the wrist joint in 32 patients (94.1%), whereas lesions were less frequently identified at the level of the forearm (n ¼ 8), upper arm (n ¼ 3), and subclavian artery (n ¼ 2). Among the 28 patients with unilateral AFI who were studied with bilateral
Table II. Arteriogram findings Finding Level of occlusive disease Hand only Forearm and hand Subclavian and hand Proximal arm, forearm, and hand Proximal arm and forearm Proximal arm only Subclavian only Incomplete palmar archa Contralateral upper extremity diseaseb Ulnar or radial aneurysm Upper arm brachial artery bifurcation
Frequency (%) 24 6 1 1 1 1 1 22 9 8 7
(68.6) (17.1) (2.9) (2.9) (2.9) (2.9) (2.9) (64.7) (32.1) (22.9) (20.0)
a
The palmar arch in one patient was not visualized on angiography. Among 28 patients with unilateral symptoms who were evaluated with bilateral upper extremity angiography. b
upper extremity angiography, nine (32.1%) had bilateral disease. Five patients underwent procedural interventions at the initial angiography, including thromboembolectomy (n ¼ 3) and subclavian artery angioplasty and stenting (n ¼ 2). Sixteen patients were treated with catheter-directed thrombolysis and studied with repeat angiography, which revealed interval anatomic improvement in eight (50%; Fig 2). Diagnostic angiography was performed without simultaneous procedural intervention or catheter-directed thrombolysis in 13 patients (37.1%), none of whom had gangrene or tissue loss at the time of presentation. Perioperative complications after angiography occurred in four patients (11.4%), including 2 access site hematomas, 1 femoral pseudoaneurysm, and 1 hematoma at the site of a peripheral intravenous catheter. The four patients with complications had been treated with catheter-directed thrombolytic infusions, and the infusions in three were discontinued after 8 to 24 hours when the complication was recognized. None of the patients with angiographyrelated complications required a blood transfusion or surgical exploration for bleeding. The femoral pseudoaneurysm occurred in a patient who was therapeutically anticoagulated with warfarin while awaiting surgical revascularization and was treated electively with ultrasound-guided thrombin injection 2 weeks after angiography. Eleven patients underwent subsequent surgical revascularization at a median of 15 days after angiography (Table III). These procedures included radial or ulnar aneurysm repair (n ¼ 7), radial or ulnar artery bypass (n ¼ 2), ulnar aneurysm excision (n ¼ 1), and radial artery thrombectomy (n ¼ 1). Adjunctive periarterial sympathectomy was performed in addition to revascularization in nine of 11 patients, and five additional patients who were not candidates for revascularization underwent periarterial sympathectomy alone. Eight patients underwent finger amputation during the follow-up period, seven of which were performed #60 days of initial presentation with tissue loss. Three of the eight finger amputations (37.5%) were performed on patients
JOURNAL OF VASCULAR SURGERY Volume -, Number -
Islam et al 5
Table III. Subsequent procedures stratified by use of catheter-directed thrombolytic infusion after initial angiographya Thrombolytic infusion Subsequent surgical procedure None Surgical revascularization þ sympathectomy Finger amputation only Sympathectomy only Surgical revascularization only Sympathectomy þ finger amputation
Yes (n ¼ 16)
No (n ¼ 19)
5 (31.3) 7 (43.8)
10 (52.6) 15 (42.9) 2 (10.5) 9 (25.7)
1 0 2 1
(6.3) (0) (12.5) (6.3)
3 3 0 1
(15.8) (15.8) (0) (5.3)
Total (N ¼ 35)
4 3 2 2
(11.4) (8.6) (5.7) (5.7)
a All procedures were performed in a delayed fashion in a separate procedural setting from angiography. Data displayed as frequency (%) by category.
whose initial arteriogram was not accompanied by simultaneous thrombectomy, endovascular intervention, or thrombolysis. Estimated 60-day amputation-free survival was 0.78 (standard error, 0.07; Fig 3). Overall amputation-free survival was similar between patients managed with vs without thrombolysis (P ¼ .61; Fig 4, A). Subgroup analysis of those patients without tissue loss or gangrene at presentation revealed a trend toward improved amputation-free survival with use of thrombolysis, with 60-day amputation-free survival of 0.92 vs 0.75 (P ¼ .12; Fig 4, B). No patients treated with surgical revascularization underwent subsequent finger amputation. At the time of last follow-up, 17 patients (48.6%) reported that they were symptom-free, 16 (45.7%) reported ongoing symptoms, and symptomatic status could not be verified in two patients. Contralateral hand symptoms developed during late follow-up in seven patients, five of whom had been evaluated with bilateral angiography and two of whom had been evaluated with unilateral angiography at the initial presentation. Symptomatic, functional, and pain scale assessments were performed at a median of 76 days, and cold sensitivity scores were performed at median of 142 days after presentation. These evaluations are summarized in Table IV and generally revealed persistent but mild symptoms. DISCUSSION AFI is a relatively uncommon condition that requires consideration of a wide variety of diagnoses, including atherosclerosis, embolization (from proximal and distal sources), vasculitis, vasospasm, thromboangiitis obliterans (Buerger disease), and thoracic outlet syndrome. Limited evidence exists to guide decision making related to evaluation and management of AFI, and the 35 patients described in this analysis represent, to our knowledge, the largest North American series reported to date. Although our analysis is limited by a retrospective design and relatively small patient cohort, we believe that the
Fig 3. Estimated overall amputation-free survival. The solid line indicates the proportion of amputation-free patients, and the vertical marks indicate censoring at time of last follow up. The number of patients at risk is indicated along the x-axis.
results provide several observations that potentially inform diagnostic evaluation and clinical management. Consistent with other previous reports, our cohort had approximately equal numbers of men and women with a high prevalence of smoking, connective tissue disorders, thrombophilia, and occupation-related repetitive hand trauma. In contrast to other reports describing a high incidence of cardioembolic AFI,11 however, only one of 35 patients in our series had a history of atrial fibrillation. Occlusive disease of the innominate or subclavian arteries, or both, was also relatively uncommon, whereas lesions distal to the wrist were identified in 32 of 35 patients. We believe that the predominately distal anatomic distribution of the pathology identified in this series supports use of angiography as the primary imaging modality for evaluation of AFI, primarily due to advantages over cross-sectional imaging for characterizing occlusive disease of the palmar or digital vessels, or both. In addition, angiography permits endovascular treatment during the same procedure when anatomically suitable disease is identified. Bilateral upper extremity arterial occlusive disease was identified in approximately one-third of patients with unilateral AFI who were studied with bilateral angiography, suggesting that AFI often occurs in the setting of chronic systemic disease that is not limited to the symptomatic extremity. Given that bilateral symptoms developed during follow-up in more than half of the patients with asymptomatic contralateral disease identified by angiography, our data also suggest that identification of contralateral disease warrants close surveillance with aggressive medical management. Additional studies are needed to determine whether a pre-emptive procedural intervention affects long-term
6 Islam et al
JOURNAL OF VASCULAR SURGERY --- 2014
Fig 4. Estimated amputation-free survival stratified by use of catheter-directed thrombolysis. A, Overall cohort (P ¼ .61). B, Patient subgroup without tissue loss or gangrene at the time of presentation (P ¼ .12).
risk of symptoms or finger amputation in the setting of asymptomatic disease. One-half of the patients treated with thrombolytic infusion had interval anatomic improvement when evaluated with repeat angiography #24 hours. Although this rate of anatomic improvement suggests that thrombolysis may improve amputation-free survival or expand anatomic options for subsequent revascularization procedures, or both, it is also important to note that thrombolysis was used in all four patients who experienced perioperative complications after angiography. Although these complications were access-site related and no patients required transfusion, these results underscore the importance of appropriate preoperative patient counseling and suggest that thrombolysis should be used selectively based on anticipated risk and potential benefit. Our results suggest that thrombolysis may be most beneficial among patients without tissue loss or digital gangrene; presumably, thrombolysis in these patients may reestablish patency of small-caliber distal arteries, which are critical for finger salvage. Accordingly, we believe that thrombolysis may be particularly useful in the setting of digital embolism, where it may reestablish patency of digital arteries poorly suited to direct surgical revascularization (such as bypass or thrombectomy, or both) due to their small diameter and potential for vasospasm with manipulation. The role of vascular reconstruction after thrombolytic therapy depends on the anatomic results and presumed etiology. Patients with radial or ulnar artery occlusion, or both, and adequate collateral circulation often improve with thrombolysis, presumably through elimination of on-going thromboembolism and reduction of associated vasospasm, before elective resection or reconstruction. In patients with inadequate collaterals and unreconstructable distal disease, thrombolysis has the potential to improve distal outflow and collateral flow, converting some patients to candidates for reconstruction and decreasing procedural complexity by reducing graft length. At our institution, reconstruction is not performed immediately to avoid the potential for residual vasospasm resulting from proximal
Table IV. Function, cold sensitivity, and pain scores (n ¼ 12)a Instrument (range of possible scores) McCabe cold sensitivity severity scale score (0-400) Levine 11-item symptom mean score (1-5) Levine 8-item function mean score (1-5) McGill pain score (visual scale scored 0-10)
Median (Q1, Q3) 254 (170, 325) 1.9 (1.5, 3.0) 1.6 (1.0, 2.4) 4.0 (0.5, 6.0)
Q1, Lower quartile; Q3, upper quartile. a Ranges for each score are displayed next to the instrument name. Symptomatic, functional, and pain scale assessments were performed at a median of 76 days, and cold sensitivity scores were performed at median of 142 days after following presentation.
arterial catheterization. In a previous report from our institution, Chloros et al12 observed a 2-year primary patency of 77% in a retrospective cohort of 13 hands in 12 patients treated with reversed interposition vein grafts for both acute and chronic disease.12 After angiography or thrombolysis, or both, for AFI in the current series, subsequent revascularization or periarterial sympathectomy, or both, were performed with greater frequency than digital amputation. Persistent long-term symptoms and functional impairment were identified in approximately half of the patients. At our center, periarterial sympathectomy is used for nonhealing digital ulcers when revascularization is not possible or as an adjunct to revascularization in patients with vasospastic disease. We prefer periarterial sympathectomy rather than central sympathectomy based on experience with this technique at our institution and the ability to combine this procedure with local revascularization or debridement, or both, under the same regional anesthetic. Patients who were not candidates for lytic treatment had no suitable targets identified, had small diameter vessels with pruning of the digital vessels, and were managed expectantly. Because pain and functional assessment scale assessments in this study were only obtained for patients
JOURNAL OF VASCULAR SURGERY Volume -, Number -
who underwent surgical revascularization or sympathectomy, or both, however, we are unable to determine how these outcomes differ in patients managed with endovascular procedures, primary amputation, or medical therapy alone. Treatment of finger ischemia with thrombolysis has been described.2-4,13 Coulon et al2 reported a series of 13 patients treated with streptokinase, and eight of 13 patients in that series had complete clot dissolution, whereas partial clearance was seen in the remainder. A high incidence of complications was also observed in their series, including six access-site hematomas, one extremity hematoma attributed to catheter-related trauma, and one transient ischemic attack. Cejna et al4 reported a retrospective analysis of 38 patients undergoing thrombolysis for acute upper extremity ischemia. These authors observed a similar anatomic disease distribution as described in this report, including 3 isolated upper arm lesions, 9 patients with combined brachial and forearm lesions, and 28 lesions located in the forearm or hand arteries, or both. Complete recanalization was achieved in proximal subclavian, axillary, and isolated brachial arterial occlusions, resulting in a success rate of 100%; however, the rate of angiographic success (defined as complete resolution of flow to the hand) was only 33%. These authors defined successful thrombolysis based on restoration of blood flow to at least one patent forearm artery supplying the palmar arch and concluded that thrombolysis results were comparable to surgical embolectomy. Miyayama et al3 described the results of catheterdirected aspiration in four patients with atrial fibrillation and upper extremity embolism, of whom three were also treated with thrombolysis. No major extremity amputations were required, and one procedure-related arterial perforation was managed with coil embolization. Several additional limitations of our analysis deserve specific mention. First, our use of a symptom duration of #30 days to define acute ischemia might be considered overly inclusive. We chose this cut point based on the general impression that, in contrast to acute lower extremity ischemia or more global upper extremity ischemia (as may result from a brachial artery embolism), AFI presentations are frequently delayed, particularly in the absence of a major neurologic deficit or gangrene. Use of a shorter interval after symptom onset might have more precisely defined the acuity of symptoms but also would have potentially excluded some patients more appropriately categorized as acute rather than chronic. Unfortunately, due to the rare nature of AFI, there is no widely accepted definition of acuity-based duration of symptoms. Second, our study excluded patients with AFI in the setting of a previous upper extremity arteriovenous hemodialysis access or major extremity trauma because distinct diagnostic considerations, management strategies, and associated outcomes for these scenarios have been described. These exclusion criteria therefore limit generalization of our results and do not inform management of patients with these diagnoses.
Islam et al 7
Third, postintervention surveillance likely resulted in longer and more detailed follow-up among patients treated with surgical or endovascular interventions, potentially biasing our results toward missed late events in patients who did not undergo revascularization procedures. Finally, we must acknowledge that patients evaluated at our hospital for AFI in departments other than ours (including those studied without catheter angiography) would have been omitted from this analysis due to our use of catheter angiography to define the study cohort. This potential selection bias might have influenced our outcomes by excluding patients with less severe symptoms or occlusive disease, or both, at the time of initial presentation. CONCLUSIONS Despite these limitations, however, this analysis provides important data related to presentation and procedural outcomes associated with AFI. Our results potentially inform decision making related to diagnostic imaging and clinical management of this uncommon problem. AUTHOR CONTRIBUTIONS Conception and design: MC, AI, ME Analysis and interpretation: AI, MC, ME Data collection: AI, CE Writing the article: AI, MC, JS Critical revision of the article: MC, AI, AK, ZL, BS Final approval of the article: MC, AI, JS, ME Statistical analysis: JS Obtained funding: Not applicable Overall responsibility: MC REFERENCES 1. Andersen LV, Mortensen LS, Lindholt JS, Faergeman O, Henneberg EW, Frost L. Upper-limb thrombo-embolectomy: national cohort study in Denmark. Eur J Vasc Endovasc Surg 2010;40:628-34. 2. Coulon M, Goffette P, Dondelinger RF. Local thrombolytic infusion in arterial ischemia of the upper limb: mid-term results. Cardiovasc Intervent Radiol 1994;17:81-6. 3. Miyayama S, Yamashiro M, Shibata Y, Hashimoto M, Yoshida M, Tsuji K, et al. Thrombolysis and thromboaspiration for acute thromboembolic occlusion in the upper extremity. Jpn J Radiol 2012;30: 180-4. 4. Cejna M, Salomonowitz E, Wohlschlager H, Zwrtek K, Bock R, Zwrtek R. rt-PA thrombolysis in acute thromboembolic upperextremity arterial occlusion. Cardiovasc Intervent Radiol 2001;24: 218-23. 5. Barbiero G, Cognolato D, Casarin A, Guarise A. Intra-arterial thrombolysis of acute hand ischaemia with or without microcatheter: preliminary experience and comparison with the literature. Radiol Med 2011;116:919-31. 6. Friedrich KM, Fruhwald-Pallamar J, Stadlbauer A, Salem G, Salomonowitz E. Hypothenar hammer syndrome: long-term follow-up of selective thrombolysis by 3.0-T MR angiography. Eur J Radiol 2010;75:e27-31. 7. Hering J, Angelkort B. [Acute ischemia of the hand after intra-arterial injection of flunitrazepam. Local combined fibrinolysis therapy in three cases]. Dtsch Med Wochenschr 2006;131:1377-80. 8. Ruch DS, Smith TL, Smith BP, Holden M, Russell G, Koman LA. Anatomic and physiologic evaluation of upper extremity ischemia. Microsurgery 1999;19:181-8.
8 Islam et al
9. Levine DW, Simmons BP, Koris MJ, Daltroy LH, Hohl GG, Fossel AH, et al. A self-administered questionnaire for the assessment of severity of symptoms and functional status in carpal tunnel syndrome. J Bone Joint Surg Am 1993;75:1585-92. 10. McCabe SJ, Mizgala C, Glickman L. The measurement of cold sensitivity of the hand. J Hand Surg Am 1991;16:1037-40. 11. Andersen LV, Mortensen LS, Lip GY, Lindholt JS, Faergeman O, Henneberg EW, et al. Atrial fibrillation and upper limb thromboembolectomy: a national cohort study. J Thromb Haemost 2011;9:1738-43.
JOURNAL OF VASCULAR SURGERY --- 2014
12. Chloros GD, Lucas RM, Li Z, Holden MB, Koman LA. Post-traumatic ulnar artery thrombosis: outcome of arterial reconstruction using reverse interpositional vein grafting at 2 years minimum follow-up. J Hand Surg 2008;33:932-40. 13. Deguara J, Ali T, Modarai B, Burnand KG. Upper limb ischemia: 20 years experience from a single center. Vascular 2005;13:84-91.
Submitted Oct 22, 2013; accepted Feb 4, 2014.