Functional outcome after infrainguinal bypass for limb salvage

Functional outcome after infrainguinal bypass for limb salvage A h m e d M. A b o u - Z a m z a m , Jr., M D , R a y m o n d W. Lee, M D , G r e g o r y L. M o n e t a , M D , L l o y d M. Taylor, Jr., M D , and J o h n M. P o r t e r , M D ,

Portland, Ore. Purpose: Functional outcome after infrainguinal bypass (IB) has recently been assessed with global health status questionnaires but not by criteria specific to the objectives of IB (i. e., maintenance of independent living and ambulation ). Preoperative and postoperative living situation and ambulatory status were evaluated in patients who underwent IB for limb salvage (LS) indications. Methods: For patients in whom IB was performed for LS from January 1980 to July 1995, living situation (independent or dependent) and ambulatory status were assessed before the onset of the need for LS surgery and 6 months after surgery. The importance of risk factors (age, sex, diabetes, heart disease, hypertension, renal insufficiency or failure, previous leg bypass, indication for surgery, postoperative morbidity, graft patency) was assessed by multivariate analysis. Results: IB for LS was performed in 513 patients. Before the development of the indication for LS surgery, 92% lived independently and 91% were ambulatory. The operative mortality rate was 2.7%. At 6 months, 86% were alive and the assisted primary graft patency rate was 92%. Ninety-nine percent of sttrvivors who lived independently before developing the need for LS surgery remained independent 6 months after surgery, and 97% of those who were ambulatory before developing the need for LS surgery were ambulatory 6 months after surgery. Only one of 25 survivors (4%) who were not living independently before surgery achieved independent living 6 months after surgery. Twenty-one percent of nonambulatory patients (6 of 29) became ambulatory. Multivariate analysis confirmed the importance of preoperative living situation and ambulatory status in predicting outcome at 6 months (p < 0.0001). Amputation and loss of primary patency were predictive of poor ambulatory status at 6 months (p < 0.0001, p --- 0.025, respectively). The overall 5-year survival rate was 48.1%. Conclusions: Preoperative independence and ambulation best predict postoperative independence and ambulation after IB for LS indications. IB procedures performed for limb salvage have a low operative mortality rate and maintain independent riving and ambulation in 99% and 97% of patients, respectively. Poor overall long-term outcome and survival in LS patients results from intercurrent illness and not from IB. (J Vase Surg 1997;25:287-97.)

O u t c o m e o f lower extremity bypass surgery has been assessed by graft patency, limb salvage, and From the Division of Vascular Surgery, Department of Surgery, Oregon Health Sciences University. Supported in part by 1RO1HL45267-01, NIH, National Heart, Lung, and Blood Institute; and MOI RR00334, General Clinical Research Centers Branch, National Institutes of Health. Presented at the Forty-fourth Scientific Meeting of the International Society for Cardiovascular Surgery, North American Chapter, Chicago, I11.,June 9-10, 1996. Reprint requests: Lloyd M. Taylor, Jr., MD, Professorof Surgery, Oregon Health Sciences University, 3181 S.W. Sam Jackson Park Road, OP- 11, Portland, OR 97201. Copyright © 1997 by The Societyfor VascularSurgery and International Society for Cardiovascular Surgery, North American Chapter. 0741-5214/97/$5.00 + 0 24/6/77980

patient survival rates. Guidelines for reporting these indicators o f surgical success have been described. 1,2 These traditional parameters, however, do not permit assessment o f patient functional status after bypass surgery. Questionnaires have been developed to assess global patient health status and functioning. One example is the Medical Outcomes Study, which developed a range o f self-reported measures to assess health status, functioning, and well-being in healthy and chronically ill patients. 3,4 A specific questionnaire has been developed and validated in patients with mild peripheral arterial disease (PAD-Walldng Impairment Questionnaire). s Currently available questionnaires focus on patient walking ability and perception of well-being. Although these are important issues in patients who 287

JOURNAL OF VASCULAR SURGERY

288

February 1997

A b o u - Z a m z a m et al.

undergo surgery for non-limb-threatening conditions, there is no validated measure of functional status available in patients with fimb-threatening ischemia. The goal of surgery for limb salvage is clearly different from that of surgery for claudication. Limb salvage surgery aims to maintain patient ambulatory status and, more importantly, potential for independent living. Patient mobility strongly correlates with perception of health. 6 Limb,salvage is preferable to primary amputation in cost as well as in allowing the patient to regain functional life. 6-11 This study was performed to assess functional outcome in patients who undergo infrainguinal bypass for limb salvage by evaluating ambulatory status and independence of fiving. METHODS

All patients who underwent infrainguinal bypass at the Oregon Health Sciences University Hospital between July 1980 and June 1995 were identified in the vascular registry. Patients with SVS grade II or III limb-threatening ischemia were chosen for further study. 2 Patients who had bilateral or repeat operations were only studied for outcome of the initial procedure to avoid repeated measures. The following patient risk factors were recorded: age, sex, indication for operation, presence of diabetes melfitus, heart disease (electrocardiographic evidence or history of ischemia, previous myocardial infarction, congestive heart failure, arrhythmia, or previous coronary bypass or angioplasty), hypertension, renal insufficiency or failure (creatinine level > 2.0 mg/dl, dialysis-dependent, or status-post renal transplant), hyperlipidemia, hypercoagulable states (routinely assessed after 1990 with protein C, protein S, antithrombin III, lupus anticoagulant, anticardiolipin antibody, resistance to activated protein C), and current or previous tobacco use. Prior vascular operative histories were documented. Intraoperative factors regarding conduit used and the quality of conduit were recorded. Perioperative morbidity (wound infection, pulmonary infection, myocardial infarction, stroke, hemorrhage) and mortality data were recorded. Primary graft patency, limb salvage, and patient survival rates were calculated. Each patient's functional status was assessed by the patient's living situation and ambulatory status before surgery (before the current indication for infrainguinal bypass arose) and 6 months after surgery. Information regarding ambulatory status and living situation was obtained by review of patient records and by direct interview of patients attending the vascular clinic during the study. Living situation was

classified as independent if patients were living alone, with a spouse, or with relatives, and as dependent if patients were living in a foster care environment or in a nursing home. Ambulatory status included independent/outside, restricted/indoors, and assisted (with cane or walker), and nonambulatory status included wheelchair dependent or bed-bound patients. Data were stored using the Dataease file management database (Dataease Int., Trumbull, Conn.). Data analysis was performed using a statistical software program (JMP 3.1, SAS Institute Inc., Cary, N.C.). Univariate analysis was by X2 analysis to identify risk factors significant for the maintenance or loss of independence in living situation or ambulatory status. A multivariate logistic regression analysis was then performed using variables found to be significant in the univariate analysis to determine factors that independently predicted favorable (independent) or unfavorable (dependent) outcomes. Primary patency, limb salvage, and survival rates were determined by standard fife-table analyses. Comparison of life-tables was by the log-rank test. All analyses were considered significant at a p value of 0.05. RESULTS

Conventional analysis During the study period, 637 patients underwent their initial infrainguinal bypass procedure for limb salvage. Of these patients, preoperative and 6-month postoperative functional status could be determined for 513 patients (81%), and these patients constitute the study population. For 124 patients (19%) preoperative or postoperative functional status, or both, could not be determined from review of the records available, and clinic visits for interviews were not possible during the study. These patients were excluded. The 124 excluded patients are compared with the 513 study patients in Tables I and II. The patients not included in the study were slightly more often male and had less heart disease. There were no other significant differences in the demographics or operative factors. The study population is therefore an appropriately representative majority sample of the entire population of patients who underwent infrainguinal bypass for limb salvage during the study period. The mean age of the study population was 66.6 years. There were 270 men and 243 women. The indication for surgery was rest pain in 229 (45%) and ischemic ulceration or gangrene in 284 (55%). Two hundred forty-eight patients (48%) had diabetes, 330 (64%) had coronary artery disease, 336 (65%) had

JOURNAL OF VASCULARSURGERY Volume 25, Number 2

Abou-Zamzam et aL

289

Table I. Patient demographics

Age (years + SD) Male Indication Rest pain Ischemic ulcer or gangrene Diabetes Heart disease Hypertension Renal insufficiency or failure Hyperlipidemia Hypercoagulable Smoking (current or previous) Previous ipsilateral bypass Procedure Fern-above-knee pop Fern-below-knee pop Fem-tibial Fern-pedal Pop-tibial Pop-pedal Other Conduit Reversed GSV Alternative vein Prosthetic

513 patients studied

124 patients with incomplete data

p

66.60 + 13.73 52.6%

68.38 ± 11.98 66.1%

0.185 0.009

44.7% 55.3% 48.3% 64.3% 65.5% 14.6% 20.5% 12.9% 72.3% 25.9%

50.0% 50.0% 57.3% 51.6% 59.7% 8.9% 12.9% 8.9% , 78.3% 30.6%

0.336 0.336 0.089 0.012 0.269 0.128 0.071 0.284 0.213 0.343

9.7% 31.2% 45.2% 3.1% 6.4% 2.7% 1.6%

6.5% 37.1% 42.7% 4.0% 5.6% 2.4% 1.6%

0.348 0.249 0.688 0.823 0.902 0.900 0.690

84.8% 11.7% 3.5%

87.9% 8.9% 3.2%

0.462 0.465 0.912

Fern, Femoral artery; pop, popliteal artery; GSV, greater saphenous vein.

hypertension, 75 (15%) had renal insufficiency or failure, and 105 (20%) had hyperlipidemia. Sixty-six patients (13%) had an identified hypercoagulable state. Three hundred seventy-one patients (72%) were current or former smokers, and 133 (26%) had undergone a previous ipsilateral leg bypass procedure (Table I). Two hundred thirty-two (45%) of the 513 procedures were femorotibial grafts, 160 (31%) were femoral-to-below-knee popliteal grafts, 50 (10%) were femoral-to-above-knee popliteal grafts, and 36 (7%) were pedal bypass grafts. The bypass conduit was autogenous vein in 495 cases (96%). Autogenous saphenous vein (374 ipsilateral, 61 contralateral) was used in 435 patients (85%), alternate vein was used in 60 (12%), and prosthetic grafts were used in 21 (4%). An adjunct inflow procedure was performed in 32 patients (6%) at the same time as the infrainguinal bypass procedure. Postoperative wound infection or failure of primary healing occurred in 60 patients (12%). Perioperative myocardial infarction occurred in 18 patients (3.5%). The operative mortality rate was 2.7%, with the majority of deaths caused by cardiac disease (Table II). The life-table assisted primary graft patency rate was 92.2% at 6 months and 72.8% at 5 years. The

Table II. Operative complications 513 patients 124 patients with studied (%) incompleie data (%) Wound infection/ delayed healing Hemorrhage Myocardial infarction Stroke Pulmonary Thrombosis Death

p

11.7

12.1

0.975

2.5 3.5 1.0 0.2 0.2 2.7

4.0 2.4 0 0 0.8 0

0.544 0.737 0.577 0.454 0.860 0.132

limb salvage rate by life table was 94.2% at 6 months and 85.2% at 5 years. The survival rate by life table was 87.8% at 6 months and 48.1% at 5 years (Table III). The patency, limb salvage, and survival rates of the majority of these procedures have been previously reported. 12 Functional outcome analysis Living situation. Before surgery 92% of patients (474 of 513) were living in an independent setting, and 8% (39 of 513) were in nursing homes. Six months after surgery, 81% of.patients (413 of 513) were living in an independent setting, 6% (30 of 513) were in nursing homes, and 14% (70 of 513) had died. Of the 474 patients who were living indepen-

290

JOURNAL OF VASCULAR SURGERY February 1997

A b o u - Z a m z a m et al.

Table III. Life-table analysis of conventional parameters Assisted primary patency rate (SEM)

Limb salvage rate (SEM)

Survival rate (SEM)

92.2% (0.01) 72.8% (0.04) 47.3% (0.11)

94.2% (0.01) 85.2% (0.03) 76.0% (0.09)

87.8% (0.01) 48.1% (0.03) 25.5% (0.05)

6-month 5-year 10-year

Table IV. Preoperative status Living situation Independent Dependent Ambulatory status Ambulatory Nonambulatory Living independently and ambulatory

Table V. Status of all patients at 6 months 92.4% 7.6% 91.4% 8.6% 87.5%

dently before surgery, 418 (88%) were alive and 56 (12%) were dead at 6 months. Ninety-nine percent of survivors (412 of 418) were able to return to independent living situations. Of the 39 patients who lived in a nursing home before surgery, 25 (64%) were alive at 6 months. Only one survivor (4%) who was in a nursing home before surgery was able to move to an independent riving situation by 6 months (Tables IV, V, and VI). Overall, living situation at 6 months was maintained in 98.4%, improved in 0.2%, and worsened in 1.4% of survivors. In the 63 patients 80 years and older who were riving independently before surgery, an independent riving situation was mainrained in 94.5% of survivors (52 of 55) at 6 months. Univariate analysis identified preoperative riving situation and ambulatory status as strong predictors of postoperative riving independence (both p < 0.001). Age greater than 80 years (p = 0.005) and the presence of ischemic ulcers/gangrene (p = 0.019) had a negative effect on living independence at 6 months. MI other variables analyzed had no predictive value (sex, diabetes, heart disease, hypertension, renal insufficiency or failure, hyperlipidemia, hypercoagulable state, smoking history, previous ipsilateral bypass, perioperative complication, primary patency, amputation). Multivariate analysis showed that preoperative independent living was independently predictive of postoperative independence in living (p < 0.0001). Preoperative ambulatory status (p = 0.76) and age 80 years or greater (p = 0.080) were not independently predictive of postoperative living situation by multivariate analysis (Table VII).

Living situation Independent Dependent Ambulatory status Ambulatory Nonambulatory Living independently and ambulatory Dead

80.5% 5.8% 79.5% 6.8% 76.4% 13.6%

Ambulatory status. Before surgery, 91% of the patients (469 of 513) were ambulatory and 9% (44 of 513) were nonambulatory. Of the ambulatory patients, 32% (149 o f 4 6 9 ) w e r e independent walkers, 46% (2!6 of 469) had some restriction, and 22% (104 of 469) required assistance or a walker (Table VIII). At 6 months, 80% of patients (408 of 513) were ambulatory, 7% (35 of 513) were nonambulatory, and 14% (70 of 513) had died. Of the 469 patients who were ambulatory before surgery, 414 (88%) were alive and 55 (12%) were dead at 6 months. Ambulatory status improved in 60% of restricted walkers who were independent walkers at 6 months, and in 36% of those who required assistance or a walker before surgery who became independent or restricted walkers (Table VIII). Ambulatory status worsened in 6% of independent walkers who became restricted or required assistance or a walker after surgery, and in 11% of restricted walkers who required assistance or a walker at 6 months after surgery (Table VIII). Ninety-seven percent of survivors who were ambulatory before surgery (402 of 414) were ambulatory at 6 months. Of the 44 patients who were nonambulatory before surgery, 29 (66%) were alive and 15 (34%) were dead at 6 months. Six of 29 survivors who were nonambulatory before surgery (21%) became ambulatory by 6 months (Tables IV, V, VI, and VIII). Overall, preoperative ambulatory status was maintained in 95.9% of survivors, worsened in 2.7% of survivors, and improved in 1.4% of survivors at 6 months. In the 67 patients 80 years and older who were ambulatory before surgery, am-

JOURNALOF VASCULARSURGERY Volume25, Number2

Abou-Zamzam et al.

Table VI. Status of survivors at 6 months Before Surgery Living situation Independent Dependent Ambulatory status Ambulatory Nonambulatory Living independently and ambulatory

6 months after surgery Independent 98.6% 4.0% Ambulatory 97.1% 20.7% Living independently and ambulatory 96.3%

Dependent 1.4% 96.0% Nonambulatory 2.9% 79.3%

bulation was maintained in 98.3% of survivors (57 of 58) at 6 months. By univariate analysis, preoperative living situation and ability to ambulate predicted postoperative ability to ambulate (both p < 0.001). Amputation (p < 0.001) and loss of primary patency (p = 0.007) were associated with loss of ambulatory status. The presence of ischemic ulcers or gangrene predicted poorer ambulatory status at 6 months (p = 0.022). Other risk factors had no predictive value (age, sex, diabetes, heart disease, hypertension, renal insufficiency or failure, hyperlipidemia, hypercoagulable state, smoking history, previous ipsilateral bypass, perioperative complication). Multivariate analysis identified preoperative ability to ambulate as the strongest independent predictor of postoperative ability to ambulate (p < 0.0001). Preoperative living situation was also predictive of ambulatory status at 6 months (p = 0.014). Amputation (p < 0.0001) and loss of primary patency (p = 0.025) predicted nonambulatory status at 6 months (Table IX). Living situation and ambulatory status. Before surgery 88% of patients (449 of 513) were both living olatside of a nursing home and ambulatory. After 6 months, 76% of patients (392 of 513) were living outside of a nursing home and ambulatory, 10% (51 of 513) were in a nursing home, nonambulatory, or both; and 14% (70 of 513) had died. Of the 449 patients independently living and ambulatory before surgery, 400 (89%) were alive and 49 (1i%) had died at 6 months. Ninety-six percent (385 of 400) of survivors were both independently living and ambulatory at 6 months (Tables IV, V, and VI). By univariate analysis, preoperative living situation and ambulatory status were strong predictors of combined status at 6 months (both p < 0.001). Amputation (p < 0.001), loss of primary patency (p = 0.004) and the presence of ischemic ulcers or gangrene (p = 0.007) predicted worse outcome at 6

291

Table VII. Predictors of living situation at 6 months

Uulvariate Preoperative living situation Preoperative ambulatory status Age ->80 years Ischemic ulcers/gangrene Multivariate Preoperative living situation Age ->80 years

<0.001 <0.001 0.005 0.019 <0.0001 0.080

months. No other risk factors appeared significant (including age, sex, diabetes, heart disease, hypertension, renal insufficiency or failure, hyperlipidemia, hypercoagulable state, smoking history, previous ipsilateral bypass, perioperative complication). Multivariate analysis showed that independence in living and walldng were strongly predictive of combined independence and ambulation at 6 months (both p < 0.0001). Amputation predicted poor combined outcome at 6 months (p = 0.0001) (Table X). Independence in 1Mng and walking were strongly related at all time points (p < 0.001). Survival. The survival rate at 6 months was 86%. The life-table 5-year survival rate was 48.1%. Univariate analysis identified preoperative nursing home residence and nonambulatory status as significant predictors of death within 6 months of operation (both p < 0.001). Renal insufficiency or failure (p < 0.001) and heart disease (p = 0.023) also predicted death within 6 months. Multivariate analysis identified nonambulatory preoperative status (p = 0.017) and renal insufficiency or failure (p < 0.0001 ) as independent predictors of death within 6 months (Table XI). Life-table analysis revealed that patients in a nursing home before surgery had a significantly poorer 5-year survival rate than non-nursing home patients (16.2% vs 50.5%, p < 0.0005). Patients with renal insufficiency or failure had a poorer 5-year survival rate than those with normal renal function (33.7% vs 50.5%, p < 0.0005). However, patients who were nonambulatory before surgery did not have a poorer 5-year survival rate than those who were ambulatory before surgery (41.9% vs 49.0%, p > 0.05) despite having a worse 6-month survival rate (70.2% vs 89.5%, p < 0.001). DISCUSSION

Numerous previous studies using conventional analysis ofpatency, limb salvage, and patient survival

JOURNAL OF VASCULAR SURGERY

292

February 1997

A b o u - Z a m z a m et al.

Table VIII. Ambulatory status 6 months after surgery Ambulatory Preoperative status

Independent 149 pts. Restricted 216 pts. Assistance 104 pts. Wheelchair 42 pts. Bed-bound 2 pts.

Independent

126 (85%) 130 (60%) 32 (31%) 2 (4.8%) 1 (50%)

Nonambulatory

Restricted

Assistance

Wheelchair

Bed-bound

4 (2.7%) 34 (16%) 5 (4.8%) 1 (2.4%) 0

5 (3.4%) 23 (11%) 43 (41%) 2 (4.8%) 0

3 (2.0%) 5 (2.3%) 4 (3.8%) 21 (50%) 1 (50%)

0 0 0 1 (2.4%) 0

Dead

11 (7.4%) 24 (11%) 20 (19.2%) 15 (36%) 0

pts., Patients.

Table IX. Predictors of ambulatory status at 6 months

Univariate Preoperative ambulatory status Preoperative living situation Amputation within 6 months Loss of primary patency Ischemic ulcers or gangrene Multivariate Preoperative ambulatory status Preoperative living situation Amputation within 6 months Loss of primary patency

<0.001 < 0.001 <0.001 0.007 0.022 <0.0001 0.014 <0.0001 0.025

data have documented that infrainguinal bypass grafting procedures performed for limb-threatening ischemia results in durable graft patency and retention of the affected limb in the majority of cases. 12-18 These same studies have also consistently documented markedly reduced long-term life expectancy in patients with lower extremity ischemia sufficiently severe to require revascularization for limb salvage. The previously reported survival rate at 5 years after surgery ranges from a low of 12%, 19 in a series of patients who were undergoing reoperation for previously failed bypass grafts, to a high of 64%. 14 The survival rate of the patients in the present series of 48% at 5 years is similar to multiple previous reports, including an earlier report of many of these same patients. 12 Despite this repeatedly documented ability of bypass surgery to achieve limb salvage, there remains legitimate concern from many physicians regarding the appropriate role of these procedures. Patients with limb-threatening ischemia are most frequently elderly (mean age in this series, 67 years; median age, 69; with 15% of patients 80 years of age or older),

and most have serious comorbid medical conditions (Table I). Because of age and medical conditions, some of these patients already have lost sufficient functional status to require nursing home placement (8% in this study), and it is reasonable to assume that many others are at significant risk. If infrainguinal bypass for limb salvage achieved limb preservation at the expense of requiring nursing home placement for a significant number of patients who previously lived independently, the overall value of the procedure would be considerably diminishe& Similarly, for many patients with limb-threatening ischemia, ambulatory function is already seriously impaired (9% in this study were nonambulatory before surgery). Limb-threatening ischemia frequently results in chronic ischemic ulcers, or areas of gangrene, the healing of which is one of the goals of limb salvage surgery. If preservation of ischemic limbs by revascularization is not accompanied by healing of wounds and preservation of ambulation, then successful limb salvage would have little meaning in functional terms. Previous attempts to quantify functional status in patients with lower extremity ischemia have used available questionnaires to assess overall health status, with the primary purpose of these studies being to describe the influence of lower extremity ischemia and its treatment on global health status. As an example, the Medical Outcomes Study (MOS) provides a means to measure clinical status, overall function, and well-being in a global health assessment? Portions of the MOS questionnaire have been applied to patients with peripheral arterial disease. 2° Questionnaires based on the MOS-Short Form 36, such as the MOS-SF20, and the RAND-SF36, as well as other global health measures described in the literature, have not been validated for assessment of patients with limb-threatening ischemia. Several re-

JOURNAL OF VASCULARSURGERY Volume 25, Number 2

A b o u - Z a m z a m et al.

Table X. Predictors of combined living situation and ambulatory status at 6 months

Univariate Preoperative living situation Preoperative ambulatory status Amputation within 6 months Loss of primary patency Ischemic ulcers or gangrene Multivariate Preoperative living situation Preoperative ambulatory status Amputation within 6 months

<0.001 <0.001 <0.001 0.004 0.007 <0.0001 <0.0001 0.0001

ports have been published using such global health questionnaires that demonstrate both the benefit of infrainguinal bypass versus amputation and improvement in global health after infralnguinal bypass. 7,8,21-23 Without validation of the questionnaire in the population being studied, the meaning of these studies is uncertain, z4,2s Attempting to assess the effect of treatment of limb-threatening ischemia using global health status questionnaires is also complicated by the very high occurrence of multiple serious medical conditions in this patient population. Improvements detected by these questionnaires coincident with treatment of limb ischemia may well reflect changes in other unrelated conditions, changes whose origin is not easily determined by the questionnaires. Although the issue of whether attempts at limb salvage are preferable to primary amputation might best be settled by a controlled study, such a study has never been performed and probably will not ever be. Certainly the present study contains no such control group. Regensteiner et al.5,26 developed and validated a questionnaire intended to assess functional status in patients with claudication, the PAD-WIQ. This valuable tool has not been validated for or applied to patients with more severe limb-threatening ischemia. Indeed, the outcome of the treatment of claudication must be viewed quite differently from that of limb-threatening ischemia. Patients with claudication have a relative impairment of walking ability, the treatment of which is elective. It is reasonable to expect such treatment to produce significant improvement in functional status to be regarded as effective. In contrast, limb-threatening ischemia results in pain at rest or gangrenous ulcers that mandate treatment. For limb-threatening ischemia, the goal of therapy is preservation of the limb and its basic function. An improvement in status is not re-

293

Table XI. Predictors of death at 6 months P Univariate Preoperative living situation Preoperative ambulatory status Renal insufficiency or failure Heart disease Multivariate Preoperative ambulatory status Renal insufficiency or failure

<0.001 <0.001 <0.001 0.023 0.017 <0.0001

quired for a successful therapeutic outcome, merely maintenance of the premorbid status quo. The present study of functional outcome after revascularization of threatened limbs was defined in terms of minimal acceptable therapeutic outcome. The parameters evaluated--independent living status and ability to ambulate--are those that, if lost, define an undesirable outcome of the surgery, independent of graft patency and preservation of the limb. Evaluation of functional results at 6 months after surgery was chosen to allow for healing of ischemic wounds and for temporary nursing home placement, which is frequently required by third-party payors as a lowercost alternative to prolonged hospitalization. At the same time, 6 months is sufficiently soon after surgery that deterioration of functional status from comorbid conditions is minimized. The authors recognize that persons who live in nursing homes are not invariably totally dependent, and those who do not live in nursing homes are not invariably independent. Many factors affect living location, including familial relations, financial resources, and concurrent illnesses. The terms dependent and independent were chosen as reasonable approximations. What is clear from this study is that limb salvage by revascularization achieves the goal of maintaining independent living (as defined by the study) and ambulation in the vast majority of those treated who survive 6 months. Ninety-six percent of survivors who were both living independently and ambulatory before surgery continued to be so 6 months after surgery. Not surprisingly, preoperative living status and ability to ambulate were the most important factors predictive of functional outcome. Sadly, the status of patients who were living in nursing homes or nonambulatory before surgery was rarely improved by limb revascularization. Only one patient moved from preoperative nursing home stares to independent living after surgery, and only 6 of 29 patients (21%) who were nonambulatory before surgery became ambulatory after surgery. Although

294

JOURNALOF VASCULARSURGERY February 1997

A b o u - Z a m z a m et al.

there may be compelling reasons to consider limb revascularization in nursing h o m e or n o n a m b u l a t o r y patients, o u r results suggest that such procedures should be considered carefully, as return to independent living or ambulation are usually n o t achievable. Clearly, this study confirms that patients w h o require limb revascularization for limb-threatening ischemia are nearing the end o f life. A l t h o u g h the operative mortality rate was acceptably low at 2.7%, an additional 10.9% had died by 6 m o n t h s after surgery, and fewer than one half (48%) survived 5 years. This shortened survival was n o t exclusively or significantly a simple function o f age. Age over 80 years was n o t predictive o f survival 6 m o n t h s after surgery, a finding in a g r e e m e n t with previously published results f r o m o u r institution. 27 I n contrast, a d v a n c e d renal insufficiency and preoperative n o n a m b u l a t o r y status were predictors o f death, e m p h a s i z i n g again that a t t e m p t s at limb salvage m u s t be carefully c o n s i d e r e d in these p a t i e n t groups. It seems unlikely t h a t the p o o r survival rate r e c o r d e d is a direct result o f the surgical treatm e n t , a l t h o u g h a r e d u c e d survival rate in patients w h o u n d e r g o elective surgical t r e a t m e n t o f claudication has b e e n suggested. 28 U n f o r t u n a t e l y , treatm e n t o f l i m b - t h r e a t e n i n g ischemia is n o t elective. Failure to revascularize leaves the patient t o face a m p u t a t i o n , a surgical t r e a t m e n t with a very welld o c u m e n t e d , very low l o n g - t e r m survival rate. 6,s,29 A l t h o u g h n o n o p e r a t i v e t r e a t m e n t s for limbt h r e a t e n i n g ischemia exist and have d e m o n s t r a t e d effectiveness in a few specific circumstances, clearly d o c u m e n t e d effectiveness for p r e v e n t i o n o f limb loss is l a c k i n g ? °,aa CONCLUSION T h e data from this study demonstrate conclusively that infrainguinal bypass grafting in patients with limb-threatening ischemia results in satisfactory functional o u t c o m e consisting o f maintenance o f ind e p e n d e n t living and ambulation in more than 96% o f surviving patients evaluated at 6 m o n t h s after surgery. T h e operative mortality rate is acceptably low at 2.7%, but the postoperative survival rate is also low, 48% at 5 years after surgery, emphasizing that limb-threatening ischemia is a condition that typically occurs in patients as they approach the end o f life. Patients w h o are n o n a m b u l a t o r y or already living in nursing h o m e s and those with advanced renal insufficiency before the occurrence o f limb-threatening ischemia infrequently experience i m p r o v e m e n t in functional status after successful revascularization, and have especially p o o r l o n g - t e r m survival rates. Infrainguinal revascularization in these patients may

still be appropriately performed after careful consideration o f individual factors, because the o u t c o m e o f available therapeutic alternatives o f amputation and nonoperative treatment is n o t d o c u m e n t e d to be any m o r e favorable. REFERENCES

1. Strandness DE, Carter SA. Outcome criteria in patients with peripheral arterial disease. Ann Vase Surg 1993;7:491-6. 2. Rutherford. RB, Flanigan DP, Gupta SK, Johnston KW, Karmody A, Whittlemore AD, et al. Suggested standards for reports dealing with lower extremity ischemia. J Vasc Surg 1986;4:80-94. 3. Stewart AL. The Medical Outcomes Study framework of health indicators. In: Stewart AL, Ware JE, editors. Measuring functioning and well-being: the Medical Outcomes Study approach. Durham, N.C.: The RAND Corporation, 1992: 12 -24. 4. Tarlov AR, Ware JE, Greenfield S, Nelson EC, Perrin E, Zubkoff M. The Medical Outcomes Study: an application of methods for monitoring the results of medical care. JAMA i989;262:92540. 5. Regensteiner JG, Steiner JF, Panzer RJ, Hiatt WIL Evaluation of walking impairment by questionnaire in patients with peripheral arterial disease. J Vase Med Biol 1990;2(3):142-52. 6. Pell JP, Donnan PT, Fowkes FGR, Ruckley CV. Quality of life following lower limb amputation for peripheral arterial disease. Eur J Vasc Surg 1993;7:448-51. 7. Paaske WP, Lanstsen J. Femorodistal bypass grafting: quality of life and socioeconomic aspects. Eur J Vasc Endovasc Surg 1995;10:226-30. 8. Thompson MM, Sayers RD, Reid A, Underwood MJ, Bell PRF. Quality of life following infragenicular bypass and lower limb amputation. Eur J Vasc Endovasc Surg 1995;9:310-3. 9. Albers M, Fratezi AC, De Luccia N. Assessment of quality of life of patients with severe ischemia as a result of infrainguinal arterial occlusive disease. J Vase Surg 1992;16:54-9. 10. Gupta SK, Veith FJ, Ascer E, Flores SA, Gleidman ML Cost factors in limb-threatening ischemia due to infrainguinal arteriosclerosis. Eur J Vasc Surg 1988 ;2:151-4. 11. Raviola CA, Nichter LS, Baker JD, Busuttil RW, Machleder HI, Moore WS. Cost of treating advanced leg ischemia: bypass graft vs primary amputation. Arch Surg 1988;123: 495-6. 12. Taylor LM, Hamre D, Dalman RL, Porter JM. Limb salvage vs amputation for critical ischemia: the role of vascular surgery. Arch Surg 1991;126:1251-8. 13. Yeager RA, Taylor LM, Porter JP. The present status of infrainguinal arterial reconstructive surgery for chronic lowerextremity ischemia. Curr Probl Surg 1991;28(2):128-39. 14. Dalman RL, Taylor L/vI. Basic data related to infrainguinal revascularization procedures. Ann Vase Surg 1990;4:309-12. 15. Mannick JA. Improved limb salvage from modern infrainguihal artery bypass techniques. Surgery 1992;111:361-2. 16. Bergamini TM, Towne JB, Bandyk DF, Seabrook GR, Schmitt DD. Experience with in situ saphenous vein bypasses during 1981 to 1989: determinant factors of long-term patency. J Vasc Surg 1991;13:137-49. 17. Donaldson MC, Mannick IA, Whittemore AD. Femoral-distal bypass with in-situ greater saphenous vein: long-term results using the Mills vaMllotome. Ann Surg 1991;213:457464. 18. Kosenbloom MS, Walsh JJ, Schuler JJ, Meyer JP, Schwarcz

JOURNALOF VASCULARSURGERY Volume25, Number2

19.

20.

21.

22.

23.

24. 25.

TH, Eldrup-Jorgensen J, et al. Long-term results of infragenicular bypasseswith autogenous vein originating from the distal superficial femoral and popliteal arteries. J Vase Surg 1988;7:691-6. Edwards JE, Taylor LM Jr, Porter JM. Treatment of failed lower extremity bypass grafts with new autogenous vein bypass grafting. J Vase Surg I990;11:136-45. Duggan MM, Woodson J, Scott TE, Ortega AN, Menzoian JO. Functional outcomes in limb salvage vascular surgery. Am J Surg 1994;168:188 91. Gibbons GW, BurgessAM, Guadagnoli E, PompaselliFB Jr, Freeman DV, Campbell DR, et al. Return to well-being and function after infrainguinal revascularization. J Vase Surg 1995;21:35-45. Schneider JR, McHorney CA, Malenka DJ, McDaniel MD, Walsh DB, Cronenwett JL. Functional health and well-being in patients with severe atherosclerotic peripheral vascular occlusive disease. Ann Vase Surg 1993;7:419 28. Hosie KB, Kockelberg R, Newbury-Ecob RA, Callum KG, Nash JR. A retrospective review of the outcome of patients over 70 years of age considered for vascular reconstruction in a district general hospital. Eur J Vase Surg 1990;4: 313-5. Testa MA, Simonson DC. Assessment of quality-of-lifeoutcomes. N Engl J Med 1996;334:835-40. Stewart AL, Greenfield S, Hays RD, Wells K, Rogers WH, Berry SD, et al. Functional status and well-being of pa-

DISCUSSION Dr. Martha D. McDaniel (Lebanon, N.H.). This work holds an important place in our growing ability to advise patients about what life will probably be like if they choose to have an operation that we are offering to them. Several surgeons in this audience have begun to investigate the best ways to achieve our main goal as surgeons, which is actually to match our care to the aims of the patients, in patients with claudication, but few have begun to answer this question for patients whose limbs are threatened. To summarize the work we have just heard: the authors recognize that our standard approach to assessment ofrevascularization outcomes does not tell the whole story from the patient's perspective. They make the presumption that we perform limb salvage revascularization operations to optimize ambulatory status and to help patients maintain or regain an independent living status. Accepting that a retrospective approach is adequate to determine this information and that the 19% of patients for whom they could not obtain information on ambulatory ability and living status are likely to be similar to those for whom they could obtain this information, the authors have demonstrated conclusively and succinctly today that in their recent experience infrainguinal revascularization procedures performed in the setting of a threatened limb do not put patients into beds and wheelchairs and do not cause them to lose independent living status if they are independent before their operation. This is an excellent first step in evaluating the efficacy of this operation that is so important in our daily working lives.

Abou-Zamzam

26.

27.

28.

29.

30.

31.

et al.

295

tients with chronic conditions: Results from the Medical Outcomes Study. JAMA 1989;262:907-13. Regensteiner JG, Hargarten ME, Rutherford RB, Hiatt WR. Functional benefits of peripheral vascular bypass surgery for patients with intermittent claudication. Angiology 1993;44: 1-10. Nehler MR, Moneta GL, Edwards JM, Yeager RA, Taylor LM Jr, Porter JM. Surgery for chronic lower extremity ischemia in patients eighty or more years of age: operative results and assessment of postoperative independence. J Vasc Surg 1993;18:618-26. Bergan JJ, Wilson SE, Wolf G, Deupree RH. Unexpected, late cardiovasculareffects of surgery for peripheral artery disease. Arch Surg 1992;127:111%24. KeagyBA, Schwartz JA, Kotb M, Burnham SJ, Johnson G Jr. Lower extremity amputation: the control series. J Vasc Surg 1986;4:321-6. RiversSP, Veith FJ, Ascer E, Gupta SK. Successfulconservative therapy of severe limb-threatening ischemia: the value of nonsympathectomy. Surgery 1986;99:75%62. Volteas N, Leon M, Labroproulos N, Christopoulos D, Boxer D, NicolaidesA. The effect ofiloprost in patients with rest pain. Eur J Vasc Surg 1993;7:654-8.

Submitted June 14, 1996; accepted Sep. 13, 1996.

As always, one good piece of work leads to more questions, and I would like to ask the authors to provide a little more information. First, you performed 45% of these procedures for rest pain and the remainder for ulceration or tissue loss. How often was the pain relieved, and how often did the focal tissue loss heal? This may be an important question inasmuch as Dr. Seabrook and his colleagues in Wisconsin found that although 53% of their limb salvage patients still had trouble with their revascularized limbs 6 months after operation, 91% reported that they were glad they had had their operations. Second, you defined three classes of patients who would qualify as ambulatory--independent and outside walkers, restricted or indoor walkers, and assisted walke r s - a n d distinguish these ambulatory patients from nonambulatory patients who needed wheelchairs or were bed-bound. You noted that deterioration in the classification into the nonambulatory group was uncommon 6 months after operation. But I wonder whether you can give us more detail about the actual walking ability among the ambulatory patients. Could they shed their canes and walk outside? And was their walking ability generally better or worse after their operations? Striking features of your data were the 50% 5-year mortality rate and the 36% 6-month mortality rate among preoperative nursing home residents. The former number is entirely consistent with our findings in metaanalyzing articles in the English language literature that deal with the outcomes of infrainguinal revasculariza-

296

A b o u - Z a m z a m et al.

tion, although one is at first shocked to hear that. Where do these results place infrainguinal revascularization procedures performed for limb salvage indications with regard to cost efficacy among the condition-treatment pairs in the Oregon ratings? We at Dartmouth are currently attracted to using a value compass for evaluating results of care. Basically, this ensures that many aspects of outcome that are important to patients are taken into consideration: clinical status, here equivalent to graft patency, limb salvage, and wound healing; functional status; cost; and patient satisfaction with results. What measures would you suggest for us to use in such a scheme for evaluating the results of limb salvage revascularization? We all appreciate the work you've done in beginning to assess functional outcome after reconstructive vascular surgery for threatened limbs and look forward to hearing your reflections on how we as a profession should proceed from here. Dr. A h m e d M. A b o u - Z a m z a m , Jr. With regard to the first question, the relief of pain, we did not directly assess the relief from the rest pain that was an indication for their operation. As far as healing, in this entire population we did not look at that in our database. We recently reported on a subgroup of these patients who underwent pedal or peroneal bypass procedures, and we found that the time to wound healing for ischemic pedal gangrene was on the average 19 weeks. So these patients obviously have a prolonged recovery time. Did the ambulatory subgroups improve? It is important to realize that what we defined as preoperative status means their ambulatory status before they developed the need for their current operation. In looking at the subgroups, more than 60% of the patients showed no change. A minority showed any change within their substrafification within the ambulatory group. As far as the question with regard to where infrainguihal bypass for limb salvage ranks in the Oregon plan, it is still now above the putative "line," so it is still comfortably covered in terms of reimbursement. I am intrigued by the idea of the value compass in assessing actual outcomes, and it sounds as though it is quite a useful tool. I think you touched on the important clinical factors, such as limb salvage, patency, patient survival, time to wound healing, and actual healing of wounds. I believe the functional status, as we describe it here, in terms of ambulation and whether the patient is living at home or needing nursing home care, is important to include in the value compass. As far as satisfaction, I think the relief of leg discomfort is a key goal. A lot of patient satisfaction has to do with their preoperative understanding of what their postoperative recovery is going to be like. We are very careful with patients to discuss that limb salvage surgery is not a cure-all and that their recovery is prolonged. With regard to future directions, I know that there are several groups worldng on how to best assess functional

JOURNAL OF VASCULARSURGERY February 1997

outcome. It's an important issue not only to vascular surgeons but to surgeons across the board, and some sort of standard measure, such as a value compass, as you described, should be developed and included in all vascular registries. I hope that addresses all your questions. Dr. John V. White (Philadelphia, Pa.). I would like to compliment you on an excellent effort here. As Dr. MeDaniel said, I think we all need to begin to look at our interventions from the patient's perspective, as you have begun t o do. Those of us who have attempted to collect data from patients recognize that asking them retrospectively to remember what type of function they had can be affected by their current psychosocial status. We also must recognize that vascular disease is a spectrum, frequently manifesting in the lower extremity as claudication, progressing to disabling claudication and, finally, to limbthreatening ischemia. The question I would pose is, as you ask the patients to reflect on their ambulatory status before the onset of indications for the need of surgery, was that before the onset of the ulcer disease, or before the onset of disabling claudication, or asking them to recollect before they had any semblance of vascular disease in the lower extremity? Dr. A b o u - Z a m z a m . We attempted to assess the patients' preoperative status just before the time that the current indications for limb salvage surgery arose. If patients had rather limiting claudication and then went on to rest pain, then their status was considered ambulatory. Not all of these patients progress slowly through the phases of claudication, rest pain, and finally tissue loss. Quite a fair number of patients who have limb-threatening ischemia will relate no clear history of claudication, and that's probably related to their comorbid conditions and their overall deconditioning. These patients aren't getting out and walking, and aren't really going to experience claudication, so ischemia becomes manifest at a late point in time. Dr. Stefan A. Carter (Winnipeg, Manitoba, Canada). I want to congratulate you on this very important and interesting work. I have some questions. First, did you stratify or look at the results in terms of the initial hemodynamic status with respect to distal circulation, such as toe pressure, oxygen tension, or some measure of pulsatility? And what was the status initially in your group? Were there some cutoff points in terms of definition or critical ischemia? In the case of renal disease, with the results both of mortality rate and function being inferior, do you recommend surgery in all those patients.,' Dr. A b o u - Z a m z a m . As far as whether we stratified these patients hemodynamically, we did not look at that in this study. The cutoff point for rest pain was an ankiebrachial index less than 0.4, and ischemic ulcers or gangrene are easily defined. As far as the results with patients who have renal insufficiency and failure, we do have a bias against operating on people who are currently undergoing hemodialysis or peritoneal dialysis. We have previously reported on our

JOURNAL OF VASCULAR SURGERY Volume 25, Number 2

results, and they have proven to be quite poor. In this current study group there are very few patients who are undergoing hemodialysis. Dr. T h o m a s S. Riles (New York, N.Y.). I certainly enjoyed your paper and its conclusions. I am a little concerned about one thing, however. The abstract I saw said that your study is based on 600-some patients, but in the presentation your calculations are all on 500 patients. It seems that there are 20% of patients who are lost to followup, which you mentioned. When you were worldng out calculations on survival rate and long-term functional results, wouldn't you suspect that those 20% of patients who were not followed might have drastically influenced the calculations that you presented to us? Do you know what happened to them? Dr. A b o u - Z a m z a m . It is a difficult situation dealing with patients with incomplete data and patients who are lost to follow-up. We deal with patients from a large geographic region who are referred by their local doctors to our institution, and so we don't necessarily have complete control over their follow-up. We try our best to get people to come in for follow-up, but we are not always successful. I think the only way to actually say for sure whether the patients lost to follow-up are similar to the ones who have been observed is to look at these patients' risk factors and their known preoperative demographics and compare

Abou-Zamzam et aL

297

them. I think that is the best way of saying that these groups are similar, but that is obviously an assumption that we can't be certain of. Dr. Gary W. G i b b o n s (Boston, Mass.). Two years ago we reported a return to function and well-being prospectively and found that the best independent predictor was how patients felt about their health and well-being at baseline, similar to what I think that you are saying. I was curious about the patients at baseline who were nonambulatory and not living independently. Was there anything about that group that would characterize a subgroup for which revascularization would not be beneficial? Dr. A b o u - Z a m z a m . I appreciate your comment. I did enjoy your paper, and I note that our results are somewhat similar. As far as this special subgroup, we did find that of patients who were both in the nursing home and nonambulatory, none of those patients achieved any improvement in their functional status. That is, thankfully, a very small proportion of our overall study group, and this subpopulation might be considered poor candidates for bypass. I think that there are a lot of other factors, such as neurologic status and the ease of care, that play into whether the patient is best served by having a functional limb on which to at least pivot, if not ambulate. We hope that studies like this will open the door for more critical questions to be answered.