Characterizing the geographic distribution of vascular surgeons in the United States.

OBJECTIVE: The shortage of vascular surgeons can be attributed to multiple factors, including an aging population, the increasing demand for vascular surgeons, and an aging vascular surgery workforce. The distribution of vascular surgeons across the United States varies by locale; thus, the shortage affects regions of different sizes disproportionately. We collated the geographic data to characterize the current distribution of vascular surgeons with an emphasis on the practice location, population density, and population age. METHODS: Vascular surgeons were identified using the Physician Compare National Downloadable file from the Centers for Medicare and Medical Services. The counties were matched with each surgeon's practice location. The locations were categorized into metropolitan, urban, or rural using the rural-urban continuum codes. Census Bureau data were used to match all counties with their population-level metrics. The distribution of vascular surgeons was analyzed by comparing the number of counties served, total patient population served, and patient population aged >50 and >65 years served. Finally, the density of vascular surgeons in the United States for the total population and for those aged >50 and >65 years was calculated. RESULTS: In 2018, the U.S. population was 309.8 million, and there were 3145 counties. Of the 3145 counties, 533 (17%) had had a practicing vascular surgeon. The combined population of these counties was 213.8 million people (69% of the U.S. population). Stratified by age, the vascular surgeons in these 533 counties could treat 37.3 million people aged >50 years and 17.4 million people aged >65 years. However, 2612 counties (83%), with a total population of 96 million people (31% of the U.S. population), had had no practicing vascular surgeon. When stratified by age, 78.1 million people in the uncovered counties were aged >50 years and 35 million were aged >65 years. Of the 2612 uncovered counties, 48% were urban and 24% were rural. CONCLUSIONS: We found a nationwide shortage of vascular surgeons, with urban and rural areas disproportionately affected negatively. Although encouraging vascular surgeons to practice in underserved areas would be an ideal solution, it is not pragmatic. Therefore, developing alternatives such as using primary care providers, investing in telehealth and developing transfer systems could be viable methods of providing vascular care to geographically isolated populations. These findings have significant implications for hospitals, patients, and vascular surgeons, who would all stand to benefit from efforts to address these disparities.

Major lower-extremity amputation: contemporary experience in a single Veterans Affairs institution.

Our objective is to describe our current experience with major lower-extremity amputation secondary to vascular disease. We conducted a retrospective review of sequential amputations over a 3-year period at one Veterans Affairs institution. One hundred thirteen amputations were performed in 99 men (age 70 +/- 11 years). Seventy-five per cent were diabetic and 23 per cent were on dialysis. Fifty-six per cent were primary amputations. The final AKA/BKA (above-knee to below-knee amputation) ratio was 3:2 and was not related to prior bypass, ethnicity, or dialysis status (P > 0.5). Forty-three per cent of amputations were BKAs in diabetics versus 26 per cent in nondiabetics (P = 0.08). The in-hospital and 30-day mortality rates were 2.6 and 8 per cent and were not related to amputation level (P = 0.76). Forty per cent experienced postoperative complications that were most frequently wound related (22%). Wound complications were more frequent with BKA than AKA (P = 0.04). At an average follow-up of 10 +/- 8 months only 65 per cent were alive. Although 51 per cent were discharged to rehabilitation units only 26 per cent regularly wore a prosthesis with 23 per cent ambulating. BKA patients were more likely to ambulate than AKA (34% vs 9%; P = 0.001), and dialysis patients were less likely to ambulate than nondialysis patients (5% vs 25%; P < 0.02). During follow-up 17 per cent of patients discharged with an intact contralateral limb required amputation of that limb and 7 per cent had bypass surgery on that limb. Complication rates were higher in African Americans and Hispanics than in whites (59%, 45%, and 23%, respectively; P < 0.001), although mortality and ambulation rates were similar. Despite an acceptable perioperative mortality complication rates remain high especially in nonwhites. One-year mortality is high. Low rehabilitation rates especially in dialysis patients mandate further efforts in this regard. Vigilant follow-up of the contralateral limb is essential.

Does the efficacy of dorsalis pedis artery bypasses vary among diabetic patients of different ethnic backgrounds?

Hispanic patients suffer from a high rate of leg amputations, far beyond what would be expected from the high prevalence of diabetes in this population. This raises questions about the efficacy of bypass operations across ethnic lines. We focused this review on dorsalis pedis bypasses, as these are frequently performed in diabetic patients. We compared outcomes between Hispanics and non-Hispanics and sought to identify factors predictive of failure or complications. The authors conducted a retrospective review of 144 dorsalis pedis bypasses in 106 men and 29 women with a mean age of 62 years. Eighty-two percent were Hispanic; 96% of cases were done for tissue loss, and 4% for rest pain. Twenty-five percent of patients experienced perioperative complications; these were more frequent in non-Hispanics than Hispanics (40% vs 22%, p = 0.05). The most frequent complications were wound related (11%). The 30-day mortality was 1.5% and 30-day graft thrombosis was 5%. Follow-up ranged from 1 to 62 months and averaged 12 months. Eighty-one percent of the limbs at risk were saved, although 36% of cases required minor foot amputations. Estimated primary graft patency was 68% at 30 months. The 30-month Kaplan-Meier curves for primary patency, assisted patency, and limb salvage were not statistically different between Hispanics and non-Hispanics (p > 0.4). Grafts that remained patent had higher duplex-derived intraoperative flow velocities in the dorsalis pedis artery than grafts that eventually failed (121 +/-69 vs 74 +/-26 cm/sec, p = 0.02). In grafts that remained patent, dorsalis pedis velocity decreased from the perioperative period to the 8 to 12 weeks time point, whereas no change was seen in grafts that eventually failed (mean decline of 48 +/-76 vs 1 +/-58 cm/sec, p = 0.02). No other factors were predictive of graft failure. The results of dorsalis pedis bypass in Hispanic patients compare favorably to those seen in other ethnic groups. This suggests that other factors must account for the high amputation rates seen in Hispanics, such as a frequent occurrence of nonreconstructible disease or unaccounted for cardiovascular risk factors. The usefulness of duplex-derived flow velocities in the dorsalis pedis to predict long-term graft patency warrants further investigation.

Does lower limb revascularization result in an improvement in sensory perception thresholds?

Patients with peripheral vascular disease are susceptible to neuropathy from chronic hypoxia. We wished to determine whether revascularization of chronically ischemic limbs results in any clinical changes in peripheral sensory thresholds. We prospectively measured quantitative vibration perception thresholds (VPT) using a Horwell neurothesiometer (in volts) in patients undergoing infrainguinal bypass, preoperatively and up to 6 month postoperatively. The bypassed limbs' preoperative VPT values were higher (lesser sensory perception) than the contralateral control limbs' preoperative values (mean score differences: 6 +/- 2 and 4 +/- 10 at toe and foot levels respectively; p ? 0.004). Preoperative VPT values were not different from the 6-month postoperative values for the revascularized legs at toe and foot levels for all 55 patients (mean score change of ?0.84 and ?1.32, p > 0.5). The contralateral limbs' VPT values did not change significantly over the 6-month period (mean change scores of 2.9, p > 0.15, and 2, p > 0.30, for toe and foot, respectively). A comparison of preoperative values between limbs that were eventually amputated and saved revealed no statistically significant differences. This study suggests that revascularization does not result in a clinically detectable improvement in sensory neuropathy. It may, however, prevent further degradation. The degree of preoperative neuropathy does not affect outcome in terms of limb salvage.

Early duplex-derived hemodynamic parameters after lower extremity bypass in diabetics: implications for mid-term outcomes.

Early postoperative changes in the hemodynamic parameters of infrainguinal bypass grafts in diabetics have not been well defined. We undertook this study to better define such changes in duplex-derived velocities and waveforms, and correlate any observed changes with intermediate-term outcomes. A prospective study of 68 primary vein bypasses for limb salvage was carried out, with scans obtained intraoperatively, daily until discharge, and at 8- to 12-weeks intervals. During follow-up (12 +/- 6 months), 20 grafts developed stenoses, 17 occluded, and 8 limbs were amputated. Most grafts show a variant of a biphasic waveform intraoperatively at the mid-graft (MG) and distal graft (DG) levels (54% and 57%); 65% of waveforms remain unchanged during the first week, and 54% remain unchanged at 3 months. No duplex-derived factors were predictive of the development of stenoses. A number of parameters were predictive of ultimate graft thrombosis. Intraoperative MG velocity was higher in grafts that eventually remained patent (83 +/- 36 vs. 60 +/- 29 cm/sec; p <0.025). Grafts that remained patent also had a much lower decline in DG and distal native (DN) velocities from immediately postoperative to 8-12 weeks later, than grafts that eventually thrombosed (-3 +/- 35 vs. -44 +/- 43 cm/sec for DG, p <0.001; and -17 +/- 66 vs. -76 +/- 53 cm/sec for DN, p <0.04 respectively). In terms of limb salvage, when the MG or DG waveform worsened (from postoperation to 12 weeks later), amputation was more likely than when it remained unchanged or improved (MG 67% vs. 9% limb loss, p <0.04; DG 43% vs. 8% limb loss, p <0.04). We conclude that intensive graft duplex surveillance does not identify grafts likely to develop stenoses. However, a number of features allow the prediction of ultimate graft failure or limb loss.

Angiographic scoring of vascular occlusive disease in the diabetic foot: relevance to bypass graft patency and limb salvage.

OBJECTIVE: We graded the severity of occlusive disease in foot vessels of patients with diabetes and correlated the scoring obtained with graft patency and limb salvage. METHODS: In this retrospective review of 199 limbs studied by means of angiography in 117 patients with diabetes mellitus, 124 limbs underwent bypass grafting. Each dorsalis pedis (DP), lateral plantar (LP), and medial plantar (MP) artery was assigned a score according to the reporting standards of the Joint Vascular Societies Council (0, no stenosis > 20%; 1, 21%-49% stenosis; 2, 50%-99% stenosis; 2.5, < half the vessel length occluded; 3, > half the vessel length occluded.) A foot score (DP + MP + LP + 1) was calculated for each foot (1 to 10). The mean follow-up period was 14 months. RESULTS: For all 199 limbs and for the 124 limbs that underwent bypass grafting, the mean scores were similar for the DP, MP, and LP (1.8 +/- 1.0, 1.9 +/- 1.0, 1.9 +/- 1.0, respectively; P >.4). Only the MP and LP correlated with each other (r = 0.57; P <.0001). There were no scoring differences between limbs with symptoms and limbs that did not undergo bypass grafting. Bypass graft patency correlated with both the foot score and the MP score for tibial and inframalleolar grafts (P <.04). Patency correlated with the LP score only for inframalleolar bypass grafting procedures. The DP score alone did not differ between bypass grafts that remained patent and bypass grafts that failed. Bypass grafts in limbs with a foot score less than 7 and an MP score less than 2 had only a 2% failure rate. A foot score greater or equal to 7 was associated with a 30% failure rate for all bypass grafts (41% for inframalleolar grafts). Bypass grafts with low foot scores that failed did so much later than bypass grafts with high foot scores that failed (17 +/- 11 months vs 6 +/- 8 months; P <.02), possibly reflecting different etiologies for the failure. The limb salvage rate correlated with foot score (P <.05). The limbs that were saved had an average foot score of 6.4 +/- 2.0, versus 7.2 +/- 1.4 for limbs that required amputation. CONCLUSION: In patients with diabetes mellitus, the foot score is a useful tool for predicting the likelihood of graft patency and limb salvage for infrapopliteal revascularization. However, the relatively high bypass success rate (70%) in the presence of a high foot score (>or= 7) does not allow its use in identifying the subgroup of patients who are unlikely to benefit from bypass grafting surgery. It cannot be used as a means of selecting patients for primary amputation.