A Prospective, Descriptive Study of Hour-to-Hour and Day-to-Day Temperature Variability of Skin Affected by Chronic Venous Disorders

  Chronic venous disorders (CVD) are the most prevalent vascular diseases, affecting an estimated 7 million adults in the US^1,2; scant epidemiologic data are available to determine the exact number of affected individuals.³ The disorders affect the deep, superficial, and perforating veins and valves of the leg with varying degrees of severity. Chronic venous disorders are complex, disabling, and costly, resulting in an estimated $1 billion to $3.5 billion in US annual expenditures for the management of the disease andtreatment of its complications.^4,5

One of the most severe complications of CVD – leg ulcer – costs, on average, $9,685 to heal.⁶ Although the actual prevalence of venous ulcers in the US is unknown, it is estimated that between 500,000 and 600,000 individuals are affected, many of whom are older adults.⁷ Individuals older than 50 years present with symptoms of greater severity, have the greatest prevalence of chronic skin ulcers (800 ulcers per 10,000 individuals), and are at highest risk of ulcer reoccurrence (more than 72%).^8,9 It is not uncommon for older adults to experience more than two venous leg ulcers over their lifetime, each with an average healing time of 7 months.¹⁰ 

  The overall prognosis of venous leg ulcers is poor; only 50% are healed at 4 months, 20% remain open at 2 years, and 8% remain open at 5 years.^7,10,11 Venous disorders account for 80% to 90% of all lower extremity ulcers, far exceeding the total number of ulcers caused by arterial insufficiency and diabetes combined.¹² Although CVD results from venous hypertension, the ulcers are believed to result from skin damage caused by inflammation and impaired microcirculation.¹³ The exact pathological mechanisms that lead from venous hypertension to skin damage and ulcers have yet to be elucidated.

  The assessment and management of CVD have been limited by the lack of sensitive and objective clinical evaluation methods, particularly for the prediction of venous ulcer development. Evidence suggests that temperature of the skin (Tsk) is elevated in the lower legs of individuals with the most severe stages of CVD-related chronic skin inflammation. The purpose of this study is to advance research in the area of infrared (IR) thermometry and CVD, such that variations in Tsk in lower legs affected by CVD and a history of leg ulcers could be tracked and assessed for trends in rhythmic patterns over multiple points in time and under real-time conditions. These data should illuminate patterns of Tsk variation in venous disease, leading to better treatment and preventive care.

  This study was designed to test the main hypothesis that Tsk variations of chronically inflamed skin of lower legs affected by CVD exhibit no differences in hour-to-hour and day-to-day rhythmic patterns associated with sleep and activities such as walking, exercise, sleeping, or compression stocking use among four selected skin sites or between the legs of individuals with CVD. A second hypothesis that the difference in temperature between sites was unequal between legs also was tested.

Literature Review

  In an exploratory study¹⁴ of 66 adults, 33 with CVD and 33 without CVD, Tsk was higher in the ankle area of affected skin compared to unaffected individuals (89.9º F ±2.7^º F versus 88.1^º P ±2.1^º F, P = 0.004). In a prospective, descriptive study¹⁵ (N = 55) of three measures of skin circulation, one of which was Tsk of CVD-affected skin (n = 31), Tsk of the medial aspect of the lower legs was examined weekly for three consecutive weeks. Skin temperature in persons with CVD was significantly (P = 0.008) higher by 2.2^º F across the three measurement periods compared to individuals without CVD (n = 24).

  Until recently, little emphasis has been accorded temperature dysregulation associated with chronic inflammation. Inflammation and impaired microcirculation are causative factors associated with skin damage prevalent in later stages of CVD. Skin temperature can be measured objectively using IR dermal thermometers.¹⁶ Due to the association with inflammation, a change in Tsk may be predictive of venous ulcer development, much like foot inflammation and elevated skin temperature are predictive of diabetic foot ulcers.¹⁷

  Inflammatory conditions affecting foot and leg temperatures have been studied with various thermometry devices such as thermistors, thermocouples, and IR thermometers^17,18 but to the authors’ knowledge, this is the first study to evaluate Tsk associated with CVD under real-time conditions using wireless thermistor technology. Wireless Tsk monitoring is a well-established biophysiological monitoring technique described elsewhere.^19,20

  Extensive research has been conducted on inflammatory processes that damage the skin, including leukocyte activation and the release of proteolytic enzymes, free radicals, metalloprotease inhibitors, adhesion molecules (such as sICAMS and sVCAMS), cytokines, and TGF-b that encourages fibroblasts to increase dermal collagen and connective tissue protein synthesis.^21-23 These inflammatory mediators and chemotactic substances lead to the formation of fibrin cuffs, scarring, tissue hypoxia, and tissue inflammation. Within the skin, microcirculation is severely compromised as a result of these inflammation-triggering events.²⁴ Together, the impaired microcirculation and inflammation elevate Tsk and increase blood flow,¹⁵ producing the clinical manifestations common to the severe stages of CVD. However, due to the spatial heterogeneity of inflammation and tissue damage, not all areas of the lower leg skin become affected with eczema, pigmentation, lipodermatosclerosis, or ulcers.

  Although guidelines and literature reviews addressing leg ulcer reoccurrence prevention are well documented,^25,26 the high recurrence rate remains a challenge to researchers and clinicians. Studies of risk factors associated with leg ulcer recurrence include history of deep vein thrombosis and lack of physical activity^27-29; however, specific clinical criteria for determining risk for ulcer reoccurrence are lacking. To date, little work has been done to establish clinical assessment parameters predictive of ulcer development.

  Increased temperature outside an individual’s typical range may be a vital sign predictive of leg ulcers. However, consensus regarding how much of a temperature increase constitutes an abnormal elevation has not been achieved; further study is required to document norms and variability of skin affected by CVD.

Methods

  Study design. A prospective, descriptive study design was used to determine Tsk variation and the ability to self-monitor Tsk among individuals with a history of leg ulcers. Skin temperature was measured in two phases in a sample of 15 subjects with CVD and a history of leg ulcers. This sample size was adequate to estimate population parameters for lower leg Tsk due to the approximately normal distribution of these data as demonstrated in pilot studies (unpublished data). After the study was approved by the Institutional Review Board, a convenience sampling method, using flyers/brochures in ambulatory care clinics, local wound clinics, senior citizen centers/residential communities, newspaper ads, and word of mouth, was used to recruit study volunteers. Once subjects were identified, provided study information, and consented to study participation via the investigator or research assistant, they were screened for inclusion/exclusion criteria.

  Inclusion/exclusion criteria. Subjects were included in the study if they demonstrated presence of venous reflux (determined by venous refilling time [VRT] of less than 25 seconds as measured with a venous Doppler); clinical manifestations of CVD with severe skin changes using clinical, etiologic, anatomic, and pathophysiologic (CEAP) staging criteria of Stage 4 (lipodermatosclerosis, hyperpigmentation, dermatitis) and Stage 5 (skin changes listed in Stage 4 plus history of healed leg ulcers); and an ankle brachial index (ABI) between 0.9 and 1.3 mm Hg (an indication of adequate arterial circulation measured with arterial Doppler). Subjects were required to demonstrate sufficient manual dexterity to use the dermal thermometer, bend slightly, and visualize leg sites. Subjects were excluded if they had an open leg ulcer; lacked venous reflux (VRT >25 sec); showed signs of arterial insufficiency (ABI <0.9 mm Hg or >1.3 mm Hg); or lacked the ability or assistance of another individual to use the thermometer, bend down to the leg, or visualize measurement sites.

  Study sample. Twenty-one (21) individuals were eligible to participate. Of those, the 15 individuals who were at highest risk of developing venous ulcers were enrolled. All had Stage 4 skin changes or Stage 5 healed venous ulcers per the CEAP classification of clinical symptoms of CVD.^30,31 Thus, all participants experienced symptoms such as eczema (redness, flakiness), lipodermatosclerosis (skin hardening), pigmentation (discoloration including brown, black, or red hues from hemosiderin deposits), atrophie blanche (lack of pigmentation over old scar tissue), and a history of leg ulcers. Skin temperature was mapped hourly over a 2-day period with a data logger and daily for 30 days with an IR thermometer.

  Variables measured. In Phase I of this study, 2 days of hourly measurements of Tsk were taken at four sites, two per leg, on each participant. Site locations were determined by IR imaging; the highest temperature readings and/or areas of healed ulcers were selected for measurement. Sites I and II were on the right leg and sites III and IV were on the left leg. In Phase II of the study, daily measurements were taken over the same four sites for 30 days. Demographic data included gender, ethnicity, family history of CVD, height, weight, ankle circumference, body mass index (BMI), compression stocking use, blood pressure, medical history, and medications.

  Data collection methods. A venous/arterial Doppler (Rheo II–Huntleigh, Watertown, NJ) was used to measure venous and arterial circulation. The IR camera (Mikron Midas Thermal Imaging Camera System, Pelham, Ala) was used to image the legs to detect elevated areas of Tsk for site measurements. A wireless temperature data logger and Tsk sensors (VitalSense Integrated Physiological Monitoring System, Mini-Mitter Co., Inc., Bend, Oreg) measured hour-to-hour Tsk in Phase I. Reported accuracy of the Tsk sensors are ± 0.1^º C (32^º to 42^º C), ± 0.25^º C (42^º to 60^º C).The TempTouch IR thermometer (Xilas Medical, San Antonio, Tex) was used by subjects to measure day-to-day skin temperature in Phase II. Accuracy reported for the IR thermometer is ±0.5^º F, with repeatability 0.2^º F. Subjects who met inclusion/exclusion criteria and provided informed consent were enrolled in the study and subsequently screened at the institution’s General Clinical Research Center (GCRC), Medical University of South Carolina, Charleston.

  The study procedures were as follows:

  A thermal image was taken of the lower extremities with the IR camera. Four areas of elevated Tsk and/or areas of previously ulcerated skin were mapped after the images were uploaded into a computer. A color contrast temperature grid was used to determine measurement sites. The areas of highest temperature only were selected for measurement. The anatomical sites with the highest temperatures (60 total: 15 subjects, two sites per leg) were included: medial in the gaitor region of leg (n = 28 sites), mid-anterior (n = 11), lateral above gaitor region (n = 9), with the remainder (12) spread throughout the entire lower leg including the backs of the legs (three). It was noted that during the image analysis, not all areas of previously ulcerated skin had higher temperatures; therefore, the protocol was amended to measure Tsk of those areas with the highest temperatures only, indicative of skin inflammation.

  For Phase I, sensors were placed over the four measurements sites, two per leg, after the skin had been prepped with alcohol to exfoliate the stratum corneum. A protective film (Skin-Prep wipe, Smith & Nephew, Largo, FL), was applied to the skin where the sensor adhesive was placed to prevent maceration or skin trauma. Cloth tape (3M, Minneapolis, MN) was applied to the outer edges of the sensor adhesive and skin to provide an additional seal.

  The data logger was activated to continuously monitor Tsk signals emitted from the sensors; the sensors, once placed during the enrollment visit, stayed on the skin for 48 hours. Study participants carried the device either in a small shoulder carrier about the size of a small camera bag or in a “fanny pack”. Participants were instructed to keep the logger in the bag at all times and to retain the bag within 3 feet of the body during rest periods, showering, bathing, or sleeping.

  All participants were quizzed on the instructions to ensure they understood treatment protocols for the 48-hour monitoring period. They also were instructed on how to contact the study principal investigator (PI) in the event of any adverse events. Follow-up phone calls were made by the researchers within 24 hours to solicit questions about study procedures and information about any problems encountered.

  Study participants returned to the GCRC after 48 hours. Sensors were removed and participants were interviewed to determine whether they encountered any problems and asked if would like to continue in the study and be enrolled in Phase II.

  For Phase II, volunteer participants were instructed on use of the IR thermometer to self-monitor temperature at home. They were instructed to take Tsk of the same four sites upon arising each morning for 30 consecutive days and record the Tsk on supplied log sheets attached to a clipboard. The sites, 1.5 cm in size, were marked with a small “x” with a skin marker. Study participants were asked to remark the sites each day, every other day, or when the “x” began to fade. They were instructed to contact the PI if a Tsk change of 2.2 C^º (4^º F) occurred. Written instructions were supplied.¹⁷

  All study participants completed a return demonstration of using the thermometer, locating measurement sites, taking Tsk at each site, and documenting data on a Tsk log sheet. Information about contacting the PI also was provided and participants were contacted by telephone 48 hours and 2 weeks after the start of Phase II to assess adherence or safety concerns. They returned to the GCRC at day 31 with the thermometer and Tsk log and completed an exit interview.

  Data analysis. In the initial descriptive analysis, the time course of Tsk for each of the four sites measured was plotted separately for each individual. The time courses were annotated to include relevant events such as awakening, exercise, or the development of an ulcer. This allowed a simple but informative visual comparison of the time courses and events of interest. Similar annotated plots were generated for the difference in Tsk between sites I and II and for the comparison of differences between sites I and II and sites III and IV. The statistical analysis of the time-series data addressed the hypothesis that Tsk does not vary according to site and/or leg. Split-plot ANOVA (mixed effects) was used to assess the significance of leg, site, and leg-site interaction effects. Least-squares means were generated for each mean (see Table 1 and Table 2). Because this study involved a small number of participants, tests of significance were performed at the a = 0.20 level.^32,33 This liberal criterion for pilot data allowed identification of trends worthy of further examination in larger samples, while not requiring the same strength of evidence necessary in a rigorous , highly-powered definitive study. All statistical analyses were completed in SAS Version 9.1 (SAS Institute, Cary, NC).

Results

  Participants. Of the 21 volunteers identified as eligible for screening, three were ineligible for the study for the following reasons: ABI <0.9 mm Hg, swelling related to chronic heart failure (not CVD), or lower leg cellulitis present before screening visit. Three other potential participants did not keep the appointment for the Phase I screening visit. Fifteen study participants completed Phase I of the study and were enrolled in Phase II. One dropped out (family crisis) during Phase II, did not return for the final visit, and did not return the temperature log. The average age of study participants was 67.7 years (range: 46.9-86.4). Most were Caucasian, had a variety of chronic conditions, and were taking medication (see Tables 3, 4, and 5).

Effect of activity on temperature.

  Phase I. The annotated time courses of hourly measurements (n = 48 measurements per subject) did not contain consistent, visually detectable effects related to caffeine use, eating, or activity. However, in 11 participants, sleeping resulted in a consistent increase in Tsk, accompanied by decreased variability (see Table 6).

  Phase II. The 48-hour and 30-day time courses (n = 30 daily measurements per subject) did not reveal any consistent trends. Figures 1 and 2 display the time courses for one subject.

  Differences between sites/legs. The mixed-effects split-plot analysis of hourly Tsk measurements for leg, site, and leg-site interaction effects identified statistical significance relevant to interaction effect (P = 0.1127). This indicates that the difference in Tsk between measurement sites is dependent on the leg on which the sites were located and that differences between sites cannot be interpreted without consideration of the leg. This confirmed the second hypothesis that the difference between sites was unequal between legs. When the interaction effect was included in the model, a significant effect for leg (P = 0.4744) or site (P = 0.5082) was not found. After accounting for leg, site, and leg-site interaction effects, the addition of BMI or ankle circumference did not result in a better-fitting model (P = 0.3997 and P = 0.8759, respectively). In Table 4, descriptive statistics are presented per subject for Phase I (N = 15). When the daily measurements were examined (Phase II, N = 14), neither leg/site nor their interaction contributed significantly to the Tsk variability. In addition, BMI (P = 0.8743), ankle circumference (P = 0.4251), and compression stocking use (P = 0.9114) did not significantly explain Tsk variation over the 30-day Phase II study period. Tables 1 and 2 describe the site means, averaged over time, for Phase I and Phase II of the study.

Discussion

  Skin temperature and blood flow are elevated in skin affected by CVD, particularly in the more severe stages of the disease in which eczema, pigmentation changes, lipodermatosclerosis, and scarring exist from previous leg ulcers. Due to impaired microcirculation, the skin becomes inflamed and is at risk for ulceration. This study investigated whether trends in Tsk could be established, such that an increase or decrease in Tsk could predict subsequent pathology such as the development of a venous leg ulcer. In the Phase I study of temperature variability of the lower leg skin affected by CVD, hourly measurements were taken over a 2-day period and participants completed an activity log to determine whether relationships between Tsk and activities such as sleeping, eating, bathing, exercise, and rest could be found. Subjects took daily measurements of the same leg sites each morning in their homes during the 30-day Phase II period.

  No detectable differences were found in the hour-to-hour Tsk measurements attributable to leg or site. No short-term variations due to caffeine use, eating, or activity were noted. However, a consistent increase in Tsk, accompanied by decreased variability, was found during periods when participants slept. This could be explained by a change in the flow of blood within the microvasculature at the skin surface during sleep that increased Tsk. These results are consistent with findings from others.³⁴ The increased Tsk was noted immediately after falling asleep and persisted until awakening. Study participants reported wearing pajamas or covering their legs with sheets/blankets, which could trap a layer of warm air between the skin and the fabric of the sheets or pajamas, altering the ambient environment.³⁵

  The increased number and chronic dilation of cutaneous capillaries associated with CVD are associated with persistent venous hypertension that apparently affects thermoregulation of the skin, even during the night.³⁶ Thus, Tsk may remain elevated over areas of inflammation. Skin blood flow produces circadian changes in body temperature; current evidence substantiates the role of Tsk in producing circadian rhythms of body temperature.^37-39 It was anticipated that Tsk would rise with exercise, eating, or drinking caffeinated beverages. These events typically trigger cutaneous vasodilation, increasing blood flow to the skin several-fold, a necessary adjustment to ensure the demand for oxygen supply to organs such as the heart is not compromised.⁴⁰ In this study population, no detectable variations were noted in response to activity. This may be due in part to inconsistent definitions and reporting of these events by study participants, which also prevented the estimation of the sleep-induced change in Tsk. The rise in Tsk could be a characteristic finding of dermatosclerotic skin, indicating that hyperperfusion in affected skin areas is a persistent problem and does not exhibit a normal circadian rhythm.²¹ If so, future research may include the development of innovative statistical models of short-term rhythmic changes in Tsk as a component of ulcer risk.

  Variation in disease severity and concomitant conditions in this heterogeneous patient group also may explain the contradictory results. Diabetes may play a role in the vasodilation/vasoconstriction responses in the skin. In this study, more than 50% of participants had type 2 diabetes. Cutaneous vasodilator dysfunction (the opening of skin blood vessels as a result of increased internal temperature) is a well-established phenomenon in diabetes even in the absence of neuropathy.

   Data suggest that patients with asymptomatic, type 2 diabetes have less vasodilation to nonpainful local warming (42^º C) compared with age-matched controls. Also, sympathetic neural control of sweating and blood pressure is significantly impaired in patients with diabetes.^37,40 These mechanisms may contribute to an impaired overall local vasodilator responsiveness, at least in the basal state, resulting in localized dysfunctional Tsk regulation.

  Several factors such as obesity and lower extremity edema also may play a role in the microcirculatory response to heat and cold. Cutaneous vasodilatory responses are more pronounced in obese individuals.⁴¹ The study population comprised primarily obese individuals – this could explain the rise in Tsk observed when the legs were covered. The effects of edema on Tsk are less clear.⁴² The effects of certain medications, such as antihypertensives that alter the sympathetic tone of the blood vessels, also should be considered, as they may raise or lower temperature. That said, the first hourly measurement taken in this study was not included in the analysis to allow the skin to acclimate after skin preparation with the alcohol and placement of the sensor. Data collection began at hour 2.

  Finally, no significant findings in the day-to-day variability of Tsk could be explained by variables such as leg, site, leg circumference, BMI, presence of diabetes, or compression stocking use. This finding suggests that the Tsk of the lower legs, at least in the morning, remains constant over time. A rise in Tsk could suggest that a pathological process such as an infection or increased inflammation is present. In one volunteer, a pathological rise of 1.5^º C at one site occurred and the skin ulcerated the next day (results are reported elsewhere).⁴³ Reports of 2.2^º C increases have been associated with the risk of developing diabetic foot ulcers.^17,18 Because of the chronic nature of skin inflammation in CVD, a rise of 1.5^º C above baseline over a 24-hour period should raise suspicion of a clinical change, although the exact increase that is a prelude to ulceration has yet to be elucidated.

  Study participants reported no problems with using the IR thermometer and expressed satisfaction with having a means to self-monitor their skin while receiving valuable feedback. One suggestion made by participants was to have a pen attached to the clipboard to prevent it from rolling off onto the floor when writing down temperatures on the temperature log. Subjects also found the temperature log to be informative – ie, they could see a visual representation of their Tsk.

Limitations

  Several study limitations have been identified that restrict the generalizability of the study findings and indicate the necessity for additional research. The sample of CVD participants was drawn from a predominantly healthcare-seeking clinical population with multiple comorbid conditions; they were taking an average of five medications each. Medications were not controlled in this study and could have influenced study results. The thermal image of the lower legs to detect “hot” spots of skin was taken at only one point in time. It is unknown whether these same sites would remain elevated from day to day compared to surrounding skin. A review of the literature revealed no previous studies of Tsk thermography of the lower legs in which multiple images were taken over an extended period of time.

  Future studies of impaired skin should incorporate repeated measures to determine variability of the Tsk over the affected sites, over time. It was also impossible to determine the extent of subject measurement error. For example, could a significant difference in Tsk occur if the subject did not measure the Tsk directly over the marked site (about 1.5 cm in size) but instead measured 2 cm distal to the site?

  In future studies, a comparison group of healthy controls would serve to establish Tsk variability in skin unaffected by vascular disorders. Also, it would be helpful to evaluate Tsk over a longer period of time to capture seasonal variations, if present.

Conclusion

  The long-term objective of this research is to determine the usefulness of IR thermometers for detecting localized Tsk elevation as a predictor for CVD skin complications such as venous leg ulcers. A risk-assessment profile for predicting venous ulcer recurrence that incorporates objective, minimally intrusive, valid, reliable, and clinically available measures of microcirculation such as Tsk remains a critical need. Leg ulcer recurrence rates associated with CVD are at an all-time high of 72% and remain an enormous clinical challenge. Treatment modalities are directed toward improving macrovascular circulation chiefly through compression, leg elevation, and exercise. Microcirculation impairments need treatment but until newer modalities are discovered, adding a self-monitoring approach to the standard of care could increase awareness of subclinical pathology. Because it has been established that Tsk in the lower legs of patients with CVD is elevated, this study sought to evaluate Tsk over two time periods to determine variations. Although this study did not establish the role of heat in predicting venous ulcers, Tsk in CVD-affected skin was not found to have a variable day-to-day pattern under predominantly basal conditions. This finding suggests that an elevation outside an individual’s range, or above one’s mean or baseline Tsk, might be indicative of occult pathology. For patients with CVD, an elevation of approximately 2.2^º C (4^º F) above baseline should serve as a warning sign, although more work needs to be done to determine a predictive temperature change marker indicating disease progression or an acute problem such as infection. Nonetheless, high-risk individuals such as those with a history of leg ulcers should be instructed to regularly monitor Tsk over areas of skin affected by CVD with eczema or redness every day or twice daily if tissue damage is suspected.

  Future research should focus on enhancing the current standard of care that specifically targets impaired skin microcirculation. Methods such as cryotherapy to reduce Tsk elevations should be considered in future studies of ulcer prevention. Individual self-monitoring of Tsk in persons affected by CVD also warrants further investigation as new advancements in IR technology make monitoring as a prediction/prevention model for venous ulcers practical and relatively inexpensive.

Acknowledgment

This research was supported by a grant from the American Nurses Foundation. Supplemental support was provided by the General Clinical Research Center #RR01070 of the Medical University of South Carolina.

1. Bergan JJ, Schmid-Schonbein GW, Smith PD, Nicolaides AN, Boisseau MR, Eklof B. Chronic venous disease. N Engl J Med. 2006;355(5):488-498.

2. Reichenberg J, Davis M. Venous ulcers. Semin Cutan Med Surg. 2005;24(4):216-226.

3. Beebe-Dimmer JL, Pfeifer JR, Engle JS, Schottenfeld D. The epidemiology of chronic venous insufficiency and varicose veins. Ann Epidemiol. 2005;15(3):175-184.

4. Rudolph DM. Pathophysiology and management of venous ulcers. J WOCN. 1998;25(5):248-255.

5. Phillips TJ, Dover JS. Leg ulcers. J Am Acad Dermatol. 1991;25(6 Pt 1):965-987.

6. Olin JW, Beusterien KM, Childs MB, Seavey C, McHugh L, Griffiths RI. Medical costs of treating venous stasis ulcers: evidence from a retrospective cohort study. Vasc Med. 1999;4(1):1-7. 7

. Burton CS. Venous ulcers. Am J Surg. 1994;167(suppl 1A):37S-41S.

8. Boccalon H, Janbon, C, Saumet JL, Faini A, Roux T, Vilain C. Characteristics of chronic venous insufficiency in 895 patients followed in general practice. Int Angiol. 1997;16(4):226-234.

9. Callum MJ, Eaglstein WH, Lynch DJ. Meeting the challenge of leg ulcer patients. Patient Care. 1988;20:24-34.

10. Fowkes FGR, Evans CJ, Lee A. Prevalence and risk factors of chronic venous insufficiency. Angiology. 2001;52(suppl 1):S5-S15.

11. Phillips T, Stanton B, Provan A, Lew R. A study of the impact of leg ulcers on quality of life: financial, social, and psychologic implications. J Am Acad Dermatol. 1994;31(1):49-53.

12. Nelzen O, Bergqvist D, Lindhagen A. Leg ulcer etiology–a cross-sectional population study. J Vasc Surg. 1991;14(4):557-564.

13. Coleridge-Smith P. The causes of skin damage and leg ulceration in chronic venous disease. Int J Lower Extremity Wounds. 2006;5(3):160-168.

14. Kelechi TJ, Haight BK, Herman J, Michel Y, Brothers T, Edlund B. Skin temperature and chronic venous insufficiency. J WOCN. 2003;3(1):17-24.

15. Kelechi TJ, Michel Y. A descriptive study of skin temperature, tissue perfusion, and tissue oxygen inpatients with chronic venous disease. Biol Res Nurs. 2007;9(1):70-80.

16. Kelechi TJ, Michel Y, Wiseman J. Are infrared and thermistor thermometers interchangeable for measuring localized skin temperature? J Nurs Meas. 2006;14(1):19-30.

17. Lavery LA, Higgins KR, Lanctot DR, et al. Preventing diabetic foot ulcer recurrence in high-risk patients: use of temperature monitoring as a self-assessment tool. Diabetes Care. 2007;30(1):14-20.

18. Lavery LA, Higgins KR, Lanctot DR, et al. Home monitoring of foot skin temperatures to prevent ulceration. Diabetes Care. 2004;27(11):2642-2647.

19. Van Marken Lichtenbelt WD, Daanen HA, Wouters L, et al. Evaluation of wireless determination of skin temperature using IButtons. Physiol Behav. 2006;88(4-5):489-497.

20. Bolton MP, McGuinnes K, Cooley J, Hayes L, Howell A. Technical note: a compact data logger for ambulatory skin temperature measurement. J Med Eng Technol. 2001;25(6):264-268.

21. Nicolaides AN. Investigation of chronic venous insufficiency: a consensus statement. Circulation. 2000:102(20):e126-e163.

22. Duran W, Pappas PJ, Schmid-Schonbien GW. Microcirculatory inflammation in chronic venous insufficiency: current status and future directions. Microcirculation. 2000;7(6 Pt 2):S49-S58.

23. Mlacak B, Blinc A, Gale N, Ivka B. Microcirculation disturbances in patients with venous ulcer before and after healing as assessed by laser Doppler flux-metry. Arch Med Res. 2005;36(5):480-484. 

24. Wollina U, Abdel-Naser MB, Mani R. A review of the microcirculation in skin in patients with chronic venous insufficiency: the problem and the evidence available for therapeutic options. Int J Low Extrem Wounds. 2006;5(3):169-180.

25. Wound Ostomy and Continence Nurses Society. Guidelines for Management of Wounds in Patients with Lower-extremity Venous Disease. Glenview, Ill: Wound Ostomy and Continence Nurses Society;2005.

26. Nelson EA, Bell-Syer SEM, Cullum NA. Compression for preventing recurrence of venous ulcers. Cochrane Database of Systematic Reviews 2000. Issue 4: CD002303. DOI: 10.1002/14651858.CD002303.

27. Nelson EA, Harper DR, Prescott RJ, Gibson B, Brown D, Ruckley CV. Prevention of recurrence of venous ulceration: randomized controlled trial of class 2 and class 3 elastic compression. J Vasc Surg. 2006;44(4):803-808.

28. Gohel MS, Taylor M, Earnshaw JJ, Heather BP, Poskitt KR, Whyman MR. Risk factors for delaying healing and recurrence of chronic venous ulcers – an analysis of 1,324 legs. Eur J Vasc Endovasc Surg. 2005;29(1):74-77.

29. Jawien A. The influence of environmental factors in chronic venous insufficiency. Angiology. 2003;54(suppl 1):S19-S31.

30. Padberg FT. CEAP classification for chronic venous disease. Dis Mon. 2005;51(2-3):176-182.

31. Eklof B, Rutherford RB, Gergan JB, et al. Revision of the CEAP classification for chronic venous disease: a consensus statement. J Vasc Surg. 2004;40(6):1248-1252.

32. Schoenfeld D. Statistical considerations for pilot studies. Int J Radiat Oncol Biol Phys. 1980;6(3):371-374.

33. Carter RE, Woolson RF. Statistical design considerations for pilot studies transitioning therapies from the bench to the bedside. J Transl Med. 2004;2:37.

34. Van Someren EJ. More than a marker: interaction between the circadian regulation of temperature and sleep, age-related changes, and treatment possibilities. Chronobiol Int. 2000;17(3):313-354.

35. Jeong WS, Tokura H. Circadian rhythm of rectal temperature in man with two different types of clothing. Int Arch Occup Environ Health. 1990;62(4):295-298.

36. Nicolaides AN. Chronic venous disease and the leukocyte-endothelium interaction: from symptoms to ulceration. Angiology. 2005;56(suppl 1):S11-S119.

37. Aldemir H, Atkinson G, Cable T, Edwards B, Waterhouse J, Reilly T. A comparison of the immediate effects of moderate exercise in the late morning and late afternoon on core temperature and cutaneous thermoregulatory mechanisms. Chronobiol Int. 2000;17(2):197-207.

38. Aoki K, Stephens DP, Johnson JM. Diurinal variation in cutaneous vasodilator and vasoconstrictor systems during heat stress. Am J Physiol Regul Integr Comp Physiol. 2001;281(2):R591-R595.

39. Thomas KA, Burr R, Wang S-Y, Lentz MJ, Shaver J. Axillary and thoracic skin temperatures poorly comparable to core body temperature circadian rhythm: results from 2 adult populations. Biol Res Nurs. 2004;5(3):187-194.

40. Charkoudian N. Skin blood flow in adult human thermoregulation: how it works, when it does not, and why. Mayo Clin Proc. 2003;78(5):603-612.

41. Wellhoner P, Rolle D, Lonnroth P, Strindberg L, Elam M, Dodt C. Laser-doppler flowmetry reveals rapid perfusion change in adipose tissue of lean and obese females. Am J Physiol Endocrinol Metab. 2006;29(5):E1025-E1030.

42. Pascarella L, Schmid-Schonbein GW, Bergan JJ. Microcirculation and venous ulcers: a review. Ann Vasc Surg. 2005;19(6):921-927.

43. Sayre EK, Kelechi TJ, Neal D. Sudden increase in skin temperature predicts venous ulcers: a case study. J Vasc Nurs. 2007;25:46-50.