Endovascular interventions may save limbs in elderly subjects with severe lower extremity arterial disease

Min-I SU, Cheng-Wei LIU

1. Division of Cardiology, Department of Internal Medicine, Taitung MacKay Memorial Hospital, Taitung Branch,China; 2. MacKay Medical College, New Taipei City, Taiwan, China; 3. Graduate Institute of Business Administration,College of Management, National Dong Hwa University, Hualien, Taiwan, China; 4. Division of Cardiology, Department of Internal Medicine, Tri-Service General Hospital Songshan Branch, National Defense Medical Center, Taipei, Taiwan,China; 5. Graduate Institute of Clinical Medicine, College of Medicine, National Taiwan University, Taipei, Taiwan,China

Endovascular intervention, such as percutaneous transluminal angioplasty (PTA),improves claudication and saves limbs of patients with severe lower extremity arterial disease (LEAD).[1]A previous study showed that the mortality among octogenarians was as high as 29%regardless of the type of intervention and that revascularization was associated with high periprocedural mortality.[2]A previous study had already shown that reconstructive surgery for elderly individuals over 80 years old resulted in a significantly higher mortality rate than that for patients between 70 and 80 years old, whereas endovascular intervention and primarily conservative treatment had comparable prognoses.[3]Consistently, another study showed that the risks of both overall and amputation-free survival were significantly lower with endovascular treatment than with bypass surgery in patients with critical limb ischemia.[4]A systematic review and meta-analysis of 27 studies (15 cohort and 12 randomized controlled trials) with 1 642 patients suggests that conservative treatment may be considered for nonreconstructable patients with critical limb ischemia;[5]however, because a high risk of bias and serious inconsistencies were found in the included studies, the meta-analysis provided low-quality evidence. In contrast, a small cohort study with 49 patients suggests that amputation improved quality of life and health status in fragile elderly individuals.[6]The choice of endovascular interventions or conservative treatments for elderly individuals with severe LEAD is still under debate.Therefore, we conducted the present study to investigate the effect of old age (age ≥ 85 years) on prognoses in patients with severe LEAD undergoing PTA compared with patients under the age of 85 years.

This was a retrospective cohort study and enrolled consecutive patients with severe LEAD who underwent PTA at our hospital between 2013/1/1 and 2018/12/31. As we previously reported, our study was an all-comer study, and we excluded only patients with a nonsalvageable limb who refused amputation surgery.[7]The study was approved by the Mackay Memorial Hospital with Institutional Review Board number (20MMHIS034e),and the board waived the informed consent requirement for the study patients. We defined our primary study outcomes as all-cause mortality, cardiac-related mortality, major adverse cardiovascular events (MACEs) and major adverse limb events(MALEs) at the one-year follow-up. MACEs were defined as the composite of nonfatal myocardial infarction, nonfatal stroke, and cardiac-related death;MALEs were defined as amputation due to a vascular event above the forefoot, acute limb ischemia and clinically driven target vessel revascularization.The therapeutic strategies were reported previously, including the timing of PTA and the medication use.[7]Because the presence of acute limb ischemia and Rutherford classification criteria were major risk factors for the study outcomes in patients with severe LEAD, we adjusted for both risk factors in the multivariate logistic regression analysis.Additionally, we showed that neutrophil-lymphocyte ratios were associated with the study outcomes in our previously report.[7]Therefore, we statistically adjusted for neutrophil-lymphocyte ratios in multivariate logistic regression analysis if the white blood cell count, neutrophil percentage or lymphocyte percentage was associated with the study outcomes in univariate logistic regression analysis. In multivariate logistic regression analysis, age as a continuous variable or age as a binary variable (age ≥85 or ＜ 85 years) was adjusted separately. We considered two-tailedPvalues of 0.05 or lower indicative of significance.

Our study cohort consisted of 222 patients with a mean age of 73.6 years (standard deviation: 11.5),and 53.6% were male. Of these patients, 25 patients(11.3%) were at Rutherford stage III, 54 patients(24.3%) were at stage IV, 130 patients (58.6%) were at stage V, and 13 patients (5.9%) were at stage VI.The older group had lower ratios of comorbidities such as diabetes mellitus (44.4%vs. 73.4%,P＜0.001) and chronic kidney diseases (17.8%vs. 41.2%,P= 0.002), but other baseline characteristics were not significantly different. The presentation of acute limb ischemia was significantly higher in the elderly group (24.4%vs. 9.6%,P= 0.012). The laboratory data did not differ significantly except that older individuals had lower values of body mass index (21.1 ± 3.6vs. 24.3 ± 4.1 kg/m2,P＜ 0.001), serum creatinine (3.8 ± 3.6vs. 1.8 ± 1.5 mg/dL,P＜0.001), and triglycerides (94.7 ± 46.6vs. 165.4 ± 128.9mg/ dL,P＜ 0.001) (shown in Table 1). With respect to the primary study outcomes, the older group had significantly higher ratios of all-cause mortality(37.8%vs. 19.2%,P= 0.016), but cardiac-related mortality was not significantly different between the older and control groups (17.8%vs. 10.2%,P=0.192); moreover, no significant association was found in MALEs (8.9%vs.16.9%,P= 0.175), although a tendency toward a significant difference was found in MACEs (26.7%vs. 14.1%,P= 0.070)(shown in Figure 1). In univariate logistic regression analyses, age as a continuous variable was associated with all-cause mortality (crude hazard ratio (cHR): 1.033, 95% CI: 1.006-1.060,P= 0.016) and in-hospital mortality (cHR: 1.056, 95% CI:1.006-1.108,P= 0.027) but not cardiac-related mortality (cHR: 1.012, 95% CI: 0.973-1.052,P= 0.559),MALEs (cHR: 0.979, 95% CI: 0.952-1.007,P= 0.146),or MACEs (cHR: 0.992, 95% CI: 0.959-1.026,P=0.636). Age ≥ 85vs. ＜ 85 years was associated with increased risks of all-cause mortality (cHR: 2.332,95% CI: 1.302-4.177,P= 0.004), MACEs (cHR: 2.138,95% CI: 1.074-4.256,P= 0.031), and in-hospital mortality (cHR: 3.694, 95% CI: 1.425-9.576,P=0.007). Borderline significance was found for cardiacrelated mortality (cHR: 2.101, 95% CI: 0.913-4.837,P= 0.081). No significant association was found regarding MALEs (cHR: 0.507, 95% CI: 0.179-1.439,P= 0.202) (shown in Table 1). In multivariate logistic regression analyses, the significant associations between age ≥ 85 years and the study outcomes became nonsignificant, including that of all-cause mortality (adjusted HR: 1.958, 95% CI: 0.937-4.090,P=0.074), cardiac mortality (adjusted HR: 1.628,95% CI: 0.607-4.366,P= 0.333), MACEs (adjusted HR: 1.350, 95% CI: 0.604-3.015,P= 0.465) and inhospital mortality (adjusted HR: 2.386, 95% CI:0.442-12.881,P= 0.312) (shown in Table 2). The original nonsignificant association between age ≥ 85years and MALEs changed after statistical adjustment for the confounders (adjusted HR: 0.141, 95% CI:0.026-0.772,P= 0.024), and age ≥ 85 years was associated with a decreased risk of MALEs compared with age ＜ 85 years.

Table 2 The association between the study outcomes and variables in logistic regression analyses.

Figure 1 The crude incidence of all-cause mortality was significantly greater in patients aged ≥ 85 vs. ＜ 85 years. The other study outcomes were comparable in the two groups.

Table 1 Baseline characteristics and laboratory data in patients aged ＜ 85 vs. ≥ 85 years.

Our study initially showed that elderly individuals aged 85 years or older had a significantly greater incidence of all-cause mortality but that cardiacrelated mortality, MALEs and MACEs did not differ significantly between older individuals and younger individuals. In our cohort study, older patients had a lower prevalence of comorbidities such as diabetes mellitus and chronic kidney diseases than younger patients, which indirectly implied a survival bias in the elderly. Chronic kidney disease was considered a risk factor in patients with LEAD,[8]although it was not associated with amputation-free survival in the comparison between endovascular interventions and conservative treatments.[9]These older patients could have longer lives because they had a lower prevalence of comorbidities of cardiovascular diseases before they developed severe LEAD; in other words, the older patients were physiologically healthier than the younger patients in our study. Age as a continuous or binary variable was not associated with all-cause mortality after we adequately adjusted for confounders in the statistical models. The major risk factors associated with all-cause mortality were Rutherford classifications, neutrophil-to-lymphocyte ratios, and alanine transaminase. A similar association can be found regarding cardiac-related mortality and MACEs, and the major risk factor associated with both outcomes was heart rate at baseline irrespective of whether age was presented as a continuous or binary variable. Age was not associated with MALEs in the univariate logistic regression analysis. Interestingly,the nonsignificant association between age and MALEs became significant after proper adjustment for the confounders and comorbidities in the multivariate logistic regression analysis. The factors associated with the decreased risks of MALEs were older age and body mass index, and the factors associated with the increased risks of MALEs in-cluded the Rutherford classification, medical history of chronic kidney diseases, and the serum values of total cholesterol and fasting glucose. As we previously explained, the older patients in our cohort were physiologically healthy compared with the younger patients, and the older patients with fewer prevalent comorbidities had a lower risk of incident MALEs. We thought that the selection bias in the elderly group explained the association of age with the decreased incidence of MALEs. We should not misinterpret age as a protective factor, but we should interpret the association between age and reduced MALEs as a good signal indicating that elderly patients still benefit from receiving endovascular intervention without the increased risks of mortality and MACEs compared with the younger patients.

Continued

Our results and interpretation may conflict with previous studies showing that conservative treatments might be noninferior to endovascular interventions.[10-12]Some investigators thought that not all patients with critical limb ischemia should undergo revascularization and focused on patient selection to avoid unnecessary procedures.[11]Another investigator showed that one-year mortality rates were as high as 40% in patients who underwent endovascular interventions or conservative treatment;an individualized therapeutic strategy combined with a shared-decision process was suggested in elderly patients with critical limb ischemia.[12]The PRIORITY registry used propensity score matching to identify 539 patients with critical limb ischemia. In this registry, one-year mortality was 44.1% in patients who received revascularization versus 49% in patients who received conservative treatment, but no significant difference in mortality rate was found.[10]Compared with patients without risk factors, the patients with 2-3 poor risk factors seemed to have an increased risk of mortality after they received surgical or endovascular revascularization, but the difference was not significant.[10]The negative risk factors included old age (age ≥ 85years in men or ≥ 90 years in women), heart failure,and wound-free resting pain.[10]In our study, most of our elderly patients were at Rutherford stage IV to VI, equal to the patients with two risk factors in the PRIORITY registry, but they still had comparable outcomes to the younger patients. The CRITISCH registry was a prospective study to develop first-line treatment options in patients with critical limb ischemia.[9]In this comprehensive study, endovascular interventions significantly decreased the risk of amputation-free survival in patients with more severe angiographic stenoses or occlusions(Trans-Atlantic Inter Society Consensus (TASC II)type C or D),[13]and age ＞74 years was not associated with an increased risk of amputation-free survival.[9]The one-year mortality rate was 19% in patients who underwent endovascular interventions or conservative treatments, but endovascular interventions increased amputation-free survival by 3%compared with conservative treatment.[9]Given the debate regarding hard outcomes in patients with severe LEAD, soft outcomes such as cost effectiveness may be used as alternatives. A study by Peters,et al.[14]included 195 patients aged over 70 years,and the authors evaluated the effect of endovascular interventions on improvement of the quality-adjusted life-years and incremental cost-effectiveness ratios compared with the effect of conservative treatment. The results indicated that performing endovascular interventions for patients with critical limb ischemia was cost effective. As optimal treatments are not always the same for elderly patients with severe LEAD,[3,9,12]we should select the appropriate treatment according to the patients’ condition and preferences. We suggest that age was not the only predictor of the prognosis in patients with severe LEAD according to our study results, and we should choose vascular interventions on the basis of the comorbidities, severity and angiographic findings of LEAD,[9]presence of ischemic wounds,[10]life expectancy and patient preferences.[12]

Continued

Continued

Continued

Selection bias was noted in the present study when we divided the patients by age. The older patients had lower ratios of comorbidities, and they seemed healthier than the controls. The selection bias of the older group partially explained the lower risks of the study outcomes in our study. Though our study was limited by the nature of the cohort study, we may still support a role of endovascular intervention for older patients with severe LEAD.Although life inevitably comes to an end, we believe that endovascular interventions can still save a limb in older patients with severe LEAD without a trade-off between limb salvage and procedural risks.

CONFLICT OF INTEREST

None.