A Cross-sectional Evaluation of the Association Between Lower Extremity Venous Ulceration and Predictive Risk Factors

Objective. This study sought to identify the association between a history of venous ulceration and demographic, vascular, and nonvascular risk factors in a healthy cohort of older adults. Patients. All patients older than 60 years impanelled within a primary care practice residing in Olmsted County, Minnesota on January 1, 2005 were enrolled. Methods. This was a cross-sectional study utilizing administrative data from an outpatient practice. The primary outcome was a previous history of venous ulceration. The predictor risk variables included demographic risk factors and comorbid health conditions. Data analysis involved univariable comparison between venous ulceration and the risk variables. The significant variables were placed in a final multivariable model. Results. The authors reviewed the records of 12,650 patients and identified 581 (4.6%) who had a history of venous ulceration. Venous insufficiency had the highest association with venous ulceration with an odds ratio of more than 900. Decubitus ulceration also had a high association with an odds ratio of 2.66 [95% CI: 1.74–4.07]. Older age, female gender, previous hospitalization, diabetes, renal insufficiency, peripheral vascular disease, congestive heart failure, depression, degenerative arthritis, peripheral neuropathy, hypothyroidism, and falls were associated with venous ulceration. Marital status, hyperlipidemia, hip fracture, chronic obstructive pulmonary disease, cancer, and dementia were not associated with venous ulceration. Conclusion. The relationship between venous insufficiency and venous ulceration appears to be very strong, as expected, given the etiology of disease. Conditions such as vascular disease and vascular risk factors were also highly associated with ulceration. Interestingly, decubitus ulceration as a risk was a novel finding.

Healthcare providers commonly encounter and treat venous ulcers in clinical practice. About 1% of the US population (estimated at 3 million people) experience venous ulceration, which results in pain and disability from both the wound and ongoing wound care.^1,2 Providers strongly emphasize edema control as the primary method of treatment for venous ulcers.³ Effective local wound management involves controlling excessive exudate, which is common in venous ulcers.⁴ Providers have effective treatment for venous ulcers; however, most clinicians and patients would prefer prevention over treatment. Venous disease experts have developed consensus statements for the prevention of venous ulcers.⁵ For successful preventative strategy implementation, providers often try to identify the patients who are at the highest risk for ulcer development. Among the older population, a stratified approach appeals to both patients and providers because of the challenges of edema treatment in frail, older adults. However, scientists in clinical research have not found these risk factors within an older population; thus, one can only draw from previous work experiences with younger populations.      

While evidence of risk factors in the older population is lacking, there are many identifiable risk factors for venous ulcers in younger cohorts. One of the most powerful predictors is a history of venous ulceration.⁶ Venous ulcers recur frequently and patients with venous insufficiency also develop venous ulceration much more frequently.⁷ In community-dwelling older adults, questions remain about the potential risks for venous ulceration understanding that previous studies had broader, younger age ranges.⁸ The objective of this study was to determine the association between demographic, vascular, and comorbid health factors in addition to the presence of venous ulceration within an older population. A cross-sectional study of all community dwelling elderly in one outpatient practice was undertaken to answer this question. The results of this study will provide important epidemiologic insights into venous ulcers in older adults and can potentially help providers stratify patients at risk for venous ulceration.

This was a cross-sectional study of adult patients who were impaneled within the division of primary care internal medicine (PCIM) at Mayo Clinic (Rochester, MN). The subjects included only patients who were older than 60 years on January 1, 2005. This academic internal medical practice included attending faculty physicians, internal medicine residents, and postgraduate geriatric fellows. The Mayo Clinic Institutional Review Board (IRB) reviewed and approved the protocol. The authors conducted all aspects of the research on this project in accordance with the principles of the Declaration of Helsinki.⁹ The investigators adhered to Minnesota state statue with regard to medical record use and privacy.¹⁰ Department of Health Science Research staff abstracted all of the information using the Mayo Clinic electronic medical record (EMR). They also abstracted information using administrative data within the electronic environment.      

Inclusion criteria. The inclusion criteria consisted of adults older than 60 years who identified a PCIM physician as their primary care provider. All subjects lived within the community or within an assisted living facility in Olmsted County, Minnesota on January 1, 2005.      

Exclusion criteria. Residents within a skilled nursing facility on January 1, 2005 were excluded from the study. Patients who did not give consent (in accordance with Minnesota state law) were excluded from analysis.      

Data collection. Health Science personnel extracted data from the EMR of the patients enrolled in the study. The study personal were blinded to the study hypothesis. Mayo Clinic Rochester maintains EMR information within one electronic system, which is readily accessible for clinical use. The authors collected predictor variables of demographics, vascular risk factors, and other comorbid risk factors. The demographic variables collected included the following: date of birth (age), gender (male or female), previous hospitalizations within the prior 2 years, and marital status. Vascular comorbid health conditions included diabetes, peripheral vascular disease, renal insufficiency, congestive heart failure (CHF), venous insufficiency, hypertension, hyperlipidemia, stroke, and combined cardiac outcomes of a history of coronary artery disease/CHF or myocardial infarction. Other comorbid medical illnesses included a history of cancer, depression, dementia, rheumatoid arthritis, degenerative arthritis, peripheral neuropathy, hypothyroidism, decubitus ulcer, fall, hip fracture, and chronic obstructive pulmonary disease (COPD). The primary outcome variable was the presence of previous venous ulceration. The diagnosis of venous ulceration was made clinically by healthcare providers and was documented in the medical record. The vascular and comorbid illnesses were likewise made clinically by the patient's primary care physician.

Data analysis. Health Research Services directly entered all information via electronic abstraction into a Microsoft Excel (Microsoft, Redmond, WA) spreadsheet for data entry, retrieval, and analysis. The investigators analyzed the final information using SAS 9.13 (SAS Institute Inc., Cary, NC). The analysis was undertaken in three parts. The initial analysis included unadjusted analysis of individual variables (demographics, vascular comorbid health concerns, and other comorbid health concerns) and the presence of previous venous ulceration by either Pearson chi-square tests or 2-sample t-tests. Predictor variables were considered significant if the two-tailed P < 0.05. After the initial unadjusted analysis, we performed an age-adjusted logistic model providing odds ratios and 95% confidence intervals when age and each predictor were considered. Both the unadjusted and the age-adjusted models were evaluated for significance for each predictor if a P < 0.05. Lastly, the multivariable logistic model was constructed using all factors with a P < 0.10. Using a stepwise approach, each variable was placed in a multivariable logistic model. Those factors in the final model with a P < 0.05 were considered significant.

A total of 12,650 patients > 60 years gave consent for medical record review. Of the 12,650 patients, 581 patients (4.6%) had previously experienced a venous ulcer. All patients with a venous ulcer had a diagnosis of venous insufficiency (100%) compared to only less than 1% in the group without a venous ulcer (P < 0.001). In subjects with venous ulceration, 5% had a history of decubitus ulceration compared to only 1% in those patients without a venous ulcer. The demographic information including the percentages or means (± standard deviation) for both patients with and without a history of venous ulceration is summarized in Table 1.      

Venous insufficiency constituted the highest odds ratio at 900 with a 95% CI of 1.44 to more than 900. Having had a previous decubitus ulcer also placed the patient at high risk, with an odds ratio of 2.66 [95% CI, 1.74-4.07]. Degenerative joint disease had an odds ratio of 2.32 [95% CI, 1.88-2.86]. All vascular risk factors except hyperlipidemia were associated with previous venous ulceration. Of the nonvascular comorbid health conditions, depression, rheumatoid arthritis, degenerative arthritis, peripheral neuropathy, hypothyroidism, and falls were all associated with previous venous ulceration. These findings are summarized in Table 2.      

The multivariable logistic regression model adjusted for all of the significant variables including demographic factors, vascular risk factors (except hyperlipidemia), and non-vascular comorbid health conditions. After adjustment, some factors were no longer significant and included: previous hospital stay, diabetes, congestive heart failure, hypertension, stroke, depression, and falls. The strength of the relationship changed for all risk factors except for venous insufficiency, which remained the strongest risk factor. After adjusting for all variables, there appeared to be a paradoxical change in the relationship with some predictor variables. The final multivariable model is noted in Table 3.

This study presents novel findings between the association of venous ulceration with venous insufficiency and decubitus ulceration in an older cohort. One might expect that comorbid health conditions are associated with venous ulceration and our study confirms this. Venous insufficiency showed the strongest association with venous ulceration with an odds ratio of more than 900 (95% CI, 1.49, > 900 [P < 0.001]). In previous studies of ischemic and venous leg ulceration, 72% of subjects had venous insufficiency as documented by ultrasound.¹¹ In the present study, pressure ulcers also had a strong association with previous venous ulceration possessing an odds ratio of 2.66 (95% CI 1.74, 4.07). This relationship between venous ulceration and decubitus ulceration changes after evaluation in the multivariable model; however, the initial univariable relationship was the second strongest predictor variable. This association has not been directly reported, yet many wound providers recognize multifactorial etiologies for chronic wounds in the lower extremity.¹² Overall, the strongest association exists between venous insufficiency and venous ulceration.      

Venous ulceration commonly develops in older adults and this study provides additional epidemiological information in this cohort. In this study of older, outpatient, community-dwelling elderly, more than 4.6% of the patients had previously suffered a venous ulcer. The incidence of venous ulceration in a broad community-based study has been estimated at 18 cases/100,000 person years.¹³ In a separate cohort study, the prevalence of venous ulceration was 0.62 per 1000 population with an increasing prevalence with age.¹⁴ In a study of hospitalized patients, the prevalence of venous ulcers was 6.54%.¹⁵ As patients develop further illnesses and eventually engage in palliative care, the prevalence of wounds increases.¹⁶ The findings in this older cohort are consistent with findings from a previous hospitalized cohort.¹⁵ Unfortunately, venous ulceration remains a costly illness in terms of direct costs and indirect expense,¹⁷ and the findings of the present study would imply that these costs would certainly affect older adults.      

The demographic risk factors for venous ulceration in our study confirm findings of risk factors in other populations. Age was significantly associated with previous venous ulceration with an odds ratio of 1.05 (95% CI, 1.04-1.06) on univariable analysis. In previous studies, age was associated with a higher prevalence of venous ulceration in both younger and older adults in a metropolitan area.¹⁴ In that study, 90% of the venous ulcers occurred in patients more than 60 years old.¹⁴ After adjustment in the multivariable model in our study, the association between age and venous ulceration actually reversed indicating a paradox. More females had venous ulcers which is consistent with a higher prevalence of venous insufficiency in women.^14,18 The incidence of venous ulceration and venous insufficiency is also higher in females in a community-based cohort.¹³ Previous hospitalization within the 2 years prior to January 1, 2005 also was associated with previous venous ulceration with an odds ratio of 1.43 (1.20-1.71). This association was not significant after multivariable modeling. However, this initial finding has not been previously mentioned in the medical literature to the authors' knowledge. Marital status maintained a protective effect for venous ulcers; however, this association became nonsignificant after adjustment.      

Beyond venous insufficiency, other vascular disease risk factors were associated with venous ulceration. Combined heart disease (coronary disease, myocardial infarction and congestive heart failure) possessed an odds ratio of 1.43 (95% CI, 1.20-1.71). Congestive heart failure (CHF) alone had an odds ratio of 1.81 (95% CI, 1.44-2.27). In the multivariable model, combined outcomes remained a risk after adjustment for other variables. Patients with CHF often suffer peripheral edema; thus, there is biological plausibility for the development of venous ulceration. Previous cohort studies of long-term care residents with venous ulcers showed an association between CHF and venous ulceration.¹⁹ Peripheral vascular disease also had a significant association with venous ulceration with an odds ratio of 1.37 (95% CI, 1.05-1.77], and this remained significant in the multivariable model. The concurrence of arterial and venous disease in lower extremity ulceration is common in many patients.^12,20 Physiologically, the lack of arterial flow in a patient with venous disease places the patient at higher risk. The remaining significant risk factors of hypertension and diabetes also are associated with venous ulceration in the univariable model. These vascular risk factors contribute to the underlying pathophysiological changes that occur with venous ulceration. Some have argued that endothelial damage and venous hypertension cause leakage of macromolecules into the intersitial space explaining the “trap” hypothesis.²¹ Diabetes and hypertension could contribute to this endovascular damage.²² In addition, diabetes can directly affect the microvascular circulation, cytokines, and other cutaneous tissue factor in the lower extremity tissues.^23,24      

Many nonvascular comorbid risk factors show some association with previous venous ulceration after adjusting for age. Degenerative arthritis was linked with venous ulceration in our study with an odds ratio of 2.32 (95% CI, 1.88-2.86). This association remained significant after adjustment. In survey studies, venous ulceration had occurred in 8.1% of patients with previous total hip replacement.²⁵ Total knee arthroplasty after severe DJD may also potentially play a role with increased edema of the lower extremity. Inflammatory arthritis also was significantly associated with venous ulceration in the present study and remained significant after adjustment. Patients with rheumatoid arthritis often suffered from venous ulceration in large epidemiological studies, which confirms our findings.²⁶ Hypothyroidism was associated with venous ulceration with an odds ratio of 1.45 (95% CI, 1.19-1.77), and remained significant in multivariable modeling. While this association has not been previously reported, this association may reflect some edema that occurs with hypothyroidism. Cancer, chronic obstructive pulmonary disease, hip fracture and dementia were not associated with venous ulceration. Cancer poses a risk for deep venous thrombosis with subsequent venous insufficiency.²⁷ Peripheral neuropathy was associated in both analysis with venous ulceration, which may reflect the underlying diabetes.²⁸ Depression was also associated with venous ulceration. Psychological conditions, specifically depression and anxiety have previously been reported in patients with venous ulceration.^29,30 The association between depression and venous ulceration does not indicate the causality. Lastly, falls were associated with venous ulcers with an odds ratio of 1.69 (95% CI, 1.37-2.09). This is a novel association between two common problems. The etiology of falls can be multifactorial; however, there may be considerable overlap in common comorbid conditions between the two illnesses.      

This study possesses numerous strengths that enhance the validity of the data. The study also suffers some inherent limitations as well. Mayo Clinic has a robust data system and comprehensive electronic medical record, which allows collection of demographic, medical utilization, and comorbid health data. There were more than 581 subjects with a previous history of venous ulceration from a community dwelling population of 12,650 patients. Administrative data systems suffer inherent weaknesses including potential missing information or miscoded information. Previous studies evaluating the accuracy of the medical record at Mayo Clinic, Rochester estimate accuracy at 97%.^31,32 A second and potentially challenging issue included those patients with venous ulceration who do not seek medical attention or treatment. These findings reflect the more serious cases of venous ulceration that seek medical attention. The ability to generalize this study outside of Olmsted County, Minnesota remains challenging in populations with vastly different outpatient populations. The population of Olmsted County is primarily Northern European-more than 90% of the population is white.³³ Despite this potential weakness, many of the risk factors in this older population confirm previous risk factors in different age and ethnic populations. Lastly, the cross-sectional design does not allow causal inference. Venous ulceration may exacerbate ongoing illnesses specifically mental health concerns such as depression.

In this older outpatient population, venous insufficiency maintained the highest level of association for venous ulceration. Most venous ulcer patients had a clinical diagnosis of venous insufficiency, putting them at the highest risk for venous ulceration. Previous decubitus ulceration was also highly associated with venous ulceration as were most of the vascular risk factors and previous vascular disease. After multivariable model adjustment, venous insufficiency remained the strongest association. From a practical standpoint, healthcare providers should assess each older individual for potential ulceration paying particularly close attention to those patients with venous insufficiency and previous ulceration (even decubitus ulceration). This study emphasizes the important association of venous ulceration in an older population.

The project described was supported by grant number 1 UL1 RR024150 from the National Center for Research Resources (NCRR), a component of the National Institutes of Health (NIH), and the NIH Roadmap for Medical Research. Its contents are solely the responsibility of the authors and do not necessarily represent the official view of NCRR or NIH. Information on NCRR is available at https://www.ncrr.nih.gov/. Information on Re-engineering the Clinical Research Enterprise can be obtained from https://nihroadmap.nih.gov.

Address correspondence to: Paul Takahashi, MD Mayo Clinic, Department of Internal Medicine 200 First St. SW Rochester, MN 55905 Phone: 507-284-2511 Email: takahashi.paul@mayo.edu