- Email: [email protected]
The relationship between pressure ulcers and skin blood flow response after a local cold provocation
40
The Relationship Between Pressure Ulcers and Skin Blood Flow Response After a Local Cold Provocation Rob J. van Marum, MD, PhD, Jan H. Meijer, PhD, Miel W. Ribbe, MD, PhD ABSTRACT. van Marum RJ, Meijer JH, Ribbe MW. The relationship between pressure ulcers and skin blood flow response after a local cold provocation. Arch Phys Med Rehabil 2002;83;40-3. Objective: To study the relationship between an impaired blood flow response after a local cold stimulus, testing nerve regulation of the local blood flow response, and an increased risk of developing pressure ulcers. Design: An observational, longitudinal, prospective study. Setting: Dutch nursing home. Patients: Eighty-two newly admitted somatic nursing home patients, age 60 years and older. Intervention: A local cold stimulus (17°C) applied to the trochanter major. Main Outcome Measures: On admission, blood flow response to a local cold stimulus. As the stimulus was withdrawn, the temperature measured at the skin increased asymptotically toward the final temperature, Tf. The velocity of this rise was characterized by the time constant, , of the process. On admission, and weekly during a 4-week follow-up period, the presence or absence of pressure ulcers was verified. Results: The blood flow response time correlated significantly with the risk of developing pressure ulcers. The patients who developed pressure ulcers during the follow-up period had a significantly longer response time than the patients who did not. Conclusions: Malfunction of the nervous regulatory mechanisms of the local blood flow is partially responsible for an increased susceptibility to pressure ulcer formation. Key Words: Autonomic nervous system; Elderly; Pressure ulcers; Rehabilitation; Skin temperature. © 2002 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation ARLY RECOGNITION OF PATIENTS at risk for developing pressure ulcers has always been an important cliniE cal problem. The risk of developing pressure ulcers results from a combination of 2 different sets of factors, the so-called external factors and internal factors.1,2 External factors are those related to the pressure (intensity, duration, direction); the internal factors (eg, fever, anemia, endothelial dysfunction, malnutrition) together determine a patient’s susceptibility to
From the Department of General Practice, Nursing Home Medicine and Social Medicine/Institute for Research in Extramural Medicine (van Marum, Ribbe); and the Department of Clinical Physics and Informatics (Meijer), Faculty of Medicine, Vrije Universiteit, Amsterdam, The Netherlands. Accepted in revised form January 11, 2001. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the author(s) or upon any organization with which the author(s) is/are associated. Reprint requests to Rob J. van Marum, MD, PhD, Dept of Geriatric Medicine, Bosch Medicentrum, PO Box 90153, ’s-Hertogenbosch, The Netherlands, e-mail: [email protected]. 0003-9993/02/8301-6480$35.00/0 doi:10.1053/apmr.2002.26827
Arch Phys Med Rehabil Vol 83, January 2002
developing pressure ulcers. This susceptibility is a patientbound characteristic that changes with time, depending on health status.1 Measuring this susceptibility has always been difficult because the underlying pathophysiologic mechanisms are unclear. Therefore, in most pressure ulcer risk assessment scores, a patient’s susceptibility to developing pressure ulcers is influenced by general physical condition, diabetes mellitus, nutritional status, and temperature.3-8 However, Meijer et al1 found that the recovery time of the local blood flow after relief of a local test pressure assessed the susceptibility to pressure ulcer formation. Susceptible patients were found to have substantially prolonged recovery times. Meijer1 calculated a patient’s risk of developing pressure ulcers by combining the outcome of the measurement of susceptibility with a measurement of the external forces acting on tissues (pressure, shear forces). The method used by Meijer,1 the so-called pressure-time-temperature (PTT) method, was based on an intense local pressure stimulus, which could be painful or induce local damage to the tissues. The objective of our study was to investigate the relationship between impaired blood flow response and increased risk for developing pressure ulcers, without using an external pressure stimulus to provoke this blood flow response. Blood flow response after pressure relief is determined by various individual mechanisms. In local autoregulation, vasoactive chemical agents formed during ischemia play an important role.1 The elastic mechanical properties of the tissue must be of importance because these properties at least determine the penetration depth of an exercised pressure, and thus the severeness of damage resulting from ischemia or distortion of the local endothelium and other tissues. The method used by Meijer,1 based on pressure as an external stimulus for the cessation of local blood flow, mainly tests local autoregulatory mechanisms. However, blood flow is also regulated by the autonomic nervous system. In a study9 of diabetic patients with autonomic neuropathy, it was suggested that impairment of the local blood flow response was the result of both endothelial damage and autonomic neuropathy. Because the main function of the autonomic vasoregulation of the skin is thermoregulation, applying a temperature stimulus can test the autonomic nervous system. The vascular response after a low-temperature stimulus is not dependent on muscle tone and mechanical deformation of the tissue, but solely on sympathetic regulation. The use of a low-temperature stimulus to provoke a local blood flow response is neither painful nor harmful to the patient. Our hypothesis was that the risk for developing pressure ulcers can be assessed by measuring the velocity of the blood flow response after a local cold provocation. METHODS Design An observational, longitudinal, prospective design was used to study the relationship between the blood flow response after a local cold stimulus and the susceptibility to the development of pressure ulcers in a cohort of newly admitted, elderly nurs-
BLOOD FLOW RESPONSE AFTER COLD PROVOCATION, van Marum
41
ing home patients. Blood flow response after a local cold stimulus was measured on admission, and the patients were examined for the existence or occurrence of pressure ulcers. During a 4-week follow-up period after admission, the patients were examined once a week for the development of new pressure ulcers. The protocol was approved by the ethics review board of the Academic Hospital of the Vrije Universiteit in Amsterdam. Subjects Data were gathered during a 34-week period from a cohort of newly admitted patients (N ⫽ 145), in a nursing home in Amsterdam, who met the following criteria: aged 60 years or older and admitted for somatic care. After application of the selection criteria, informed consent was obtained from each patient by the nursing home physician, who also assessed the patient’s suitability for participation on the basis of medical and ethical considerations (patients had to be able to maintain a horizontal position, lying on their side for at least 30min). If measurements could not be taken within 48 hours after admission, the patient was excluded. During the study period, patients could also be excluded for the following reasons: a substantial change in health status (these changes were considered to result in a health status that was no longer comparable with that on admission), discharge or death within 2 weeks after admission without developing pressure ulcers (participation in the study was then considered to be too short to observe the occurrence of pressure ulcers at the different intervals), and a pressure ulcer on admission. It was not possible to allocate a patient with pressure ulcers on admission to the group with newly developed pressure ulcers or to the group with no newly developed ulcers because stabilization or cure of the ulcers may well have been the result of extra efforts of the (para-) medical and nursing staff, which is difficult to quantify. Measurement of Blood Flow Response One task of the sympathetic nerve system is to regulate peripheral blood flow in response to temperature loads. Because damage of (peripheral) autonomic nerve fibers occurs in several diseases, widespread use has been made of temperature provocation tests to detect and assess the extent of this damage, and also to evaluate the effect of therapies on, for instance, diabetes mellitus, Raynaud’s disease, and anorexia nervosa.10-16 In these tests, local temperature stimuli (mainly cold) are applied locally, and local vascular responses are studied, usually by monitoring the local temperature course. The stimuli are usually applied to an extremity. In our study, a modified version of the method described by Hansteen15 and Cleophas et al17 was used, that is, the stimulus was not applied to the fingers but to the region of the greater trochanter. This site was chosen because it has been found to be a site at major risk for the development of pressure ulcers and that has been studied frequently. The regulatory functioning of the local sympathetic nerve system was tested by applying a local, low-temperature stimulus. The low-temperature stimulus was applied to the trochanter major region. Because patients had to keep their position for at least 30 minutes, they could choose the measurement side (left or right trochanteric region). The stimulus was applied for 7 minutes, by means of a cylindric probe with a contact surface area of 5.6cm2, through which water at a temperature of 17°C was pumped. The temperature measurements and signal processing were performed in a manner that has been described.18 The only difference in this study was that the temperature sensora was not potted into an indenter, but directly attached to
Fig 1. Example of registration representing a slow response after relief of the low-temperature stimulus. The cold stimulus is applied at T0 and removed at T1. The discrete steps in the temperature curves originate from the analog-to-digital conversion. Each step is 0.1°C.
the skin with adhesive tape. The measurements were performed in a temperature-controlled room. While lying on their side, patients were covered with a blanket, leaving 1 trochanter major exposed. They were then allowed to stabilize in the measurement position and to adapt to the environment during a period of 15 minutes before measurement. A physicist, who was unaware of the medical status of the patient, performed the measurements. Corresponding with the PTT measurements by Meijer et al,1 in which pressure was used as an external stimulus for provocation of the blood flow response, the method using temperature as a stimulus can be named the temperature-temperature-time (TTT) method. Figure 1 shows a typical example of registration. The initial skin temperature is T0. As the stimulus is applied, the skin temperature drops toward a new equilibrium (descending part of the curve). This process is dominated by the stimulus, and was not taken into account in the analysis. As the stimulus was withdrawn, the temperature measured at the skin increased asymptotically toward the final temperature, Tf. The velocity of this rise can be characterized by the time constant, , of the process ( ⬇ 1.44 times the half-life, t1/2, of the process). Both and Tf are calculated by means of exponential regression analysis, in which an exponential function is fitted to the ascending part of the curve. Although the amplitude of the response is considered to be dominated by the amplitude of the stimulus, and to contain no clinically relevant information, the difference in final temperature and initial temperature (Tf ⫺ T0) may be clinically relevant. In this study, both and Tf ⫺ T0 were included as independent variables. Measurement of Health Condition Health condition was operationalized according to the Norton score.4 A change in Norton score of more than 3 points during the follow-up period was considered to indicate a substantial change in health condition. The Norton score of each patient was routinely measured by the research physician on admission and after 4 weeks, or if necessary more frequently during the study period. Measurement of Pressure Ulcers The examinations to determine the existence of pressure ulcers during the study period were performed by a research Arch Phys Med Rehabil Vol 83, January 2002
42
BLOOD FLOW RESPONSE AFTER COLD PROVOCATION, van Marum
Table 1: Mean Response Time and Difference Between Final and Initial Skin Temperature for Patients With and Without Pressure Ulcers After 4 Weeks Patient Group
(min)
Tf ⫺ T0 (°C)
Group D⫹ (n ⫽ 21) Group D⫺ (n ⫽ 61) P
4.5 ⫾ 1.2 3.7 ⫾ 0.7 ⬍.0005
⫺1.6 ⫾ 0.8 ⫺1.4 ⫾ 0.6 NS
NOTE: Values are mean ⫾ SD. Abbreviation: NS, not significant.
physician who was unaware of the results of the measurement of the blood flow response. The first examination was performed within 48 hours after admission to the nursing home. Each patient was examined from head to toe for the presence or absence of pressure ulcers. A pressure ulcer was defined as any degenerative change, caused under influence of pressure and shear forces acting on biologic tissues. In this study, nonblanching hyperemia was considered to be the initial manifestation of a pressure ulcer. Analysis The mean response time () and the difference between final and initial temperature measurement (Tf ⫺ T0) of the group of patients with newly developed pressure ulcers (ulcers developed during follow-up period regardless of location) were compared with those of the group of patients with no newly developed pressure ulcers by means of Student’s t test. The relationship between the time difference in skin temperature, the response-time, and the risk of developing pressure ulcers was calculated by means of logistic regression analysis. These relationships were adjusted for age, gender, and prevention given. RESULTS Of the 145 patients who met the inclusion criteria, 29 patients were not measured for the following reasons: refusal (n ⫽ 6), assessed as unsuitable by the nursing home physician for medical or ethical reasons (n ⫽ 13), or measurement not performed within 48 hours after admission (n ⫽ 10). Of the remaining 116 patients who were measured, 34 patients (29%) were excluded for the following reasons: the presence of a pressure ulcer on admission that remained stable or was cured during the follow-up period (n ⫽ 20), discharge or death within 2 weeks after admission without developing a pressure ulcer (n ⫽ 9), and substantial change in health condition (n ⫽ 5). Data on 82 patients were included in the analyses. In total, 21 patients (26%) developed a pressure ulcer during the study period. Blood flow responses of patients in the group who developed pressure ulcers (D⫹ group) were compared with the responses of those who did not, develop pressure ulcers (D⫺ group) (table 1). D⫹ group patients showed a significantly higher time constant than D⫺ group patients. No differences between the groups could be found with regard to the difference between final and initial skin temperature. The relationship between (range, 130 – 436s; mean ⫾ standard deviation [SD], 235 ⫾ 53s) and the risk (R) of developing pressure ulcers was significant (P ⫽ .006). This relationship showed no significant change when prevention was included as a confounder, and can be described with the formula: y ⫽ ⫺8.23 ⫹ .0159, in which y ⫽ loge(Pulcer /1 ⫺ Pulcer), where Pulcer ⫽ pressure ulcer. For each value of y, R can be calculated: R ⫽ ey (1 ⫹ ey). Arch Phys Med Rehabil Vol 83, January 2002
No significant relationship was found between Tf ⫺ T0 and the risk of developing pressure ulcers. DISCUSSION The objective of this study was to investigate whether a relationship exists between the blood flow response after a local cold stimulus, using the TTT method, and the risk for developing pressure ulcers. In this prospective evaluation, a significant relationship was found between (the response time of the skin temperature after relief of the low-temperature stimulus) and the risk for developing pressure ulcers. The D⫹ group showed a significantly slower response time compared with the D⫺ group (table 1). From these observations, it can be concluded that malfunction of the nervous regulatory mechanisms of the local blood flow is at least partially related to increased susceptibility to pressure ulcer formation. These findings support the observations reported by Mein et al,19 who investigated the influence of spinal anesthesia on the local blood flow response as a model for autonomic neuropathy. They reported a doubling of the blood flow response time after relief of a pressure stimulus during spinal anesthesia. However, in their study, no distinction could be made between the influence of blockage of the autonomic nerves and the motor nerves, resulting in paralysis of muscle tissue. Moreover, the stimulus used in their study (local pressure) induced tissue strain.19 Because in our study no tissue strain was induced, it can be concluded that the delayed blood flow response, which was also reported by Mein,19 has to be attributed to malfunction of the autonomic nervous system. The studies performed by Meijer1 and Mein,19 based on a pressure stimulus, show a much greater variation in the blood flow response, compared with the relatively small variation observed in the response provoked by a temperature stimulus, whereas both responses correlated with the risk of developing pressure ulcers. This may imply that both responses are related, at least partially, to different sets of subfactors of susceptibility. However, the reliability of the 2 methods may differ, so this assumption should be made with caution. It cannot be excluded that the experimental setting influenced the variations. There is an important difference between the pressure stimulus test and the temperature stimulus test. With the pressure stimulus, at the end of the stimulation period there is thermal equilibrium in the tissue, whereas with the temperature stimulus, there is an absence of thermal equilibrium. This implies that the velocity of the temperature response after relief of the temperature stimulus is modulated by the thermal diffusion flux, which complicates comparison of the responses of the 2 methods. The thermal diffusion flux may well be influenced by the amount of subcutaneous fat tissue. A substantial variation in the difference between final and initial skin temperature (Tf ⫺ T0) was observed. The skin temperature did not return to its original value after relief of the temperature stimulus in either group of patients. These observations suggest a persisting vasoconstriction. However, no relationship with the risk for developing pressure ulcers could be established. A similar, substantial variation was found in a previous study10 based on a comparable cold provocation test, in which fingers were cooled. However, the reported possible influence of ambient temperature on Tf ⫺ T0 does not apply to the results of our study because room temperature was kept at a minimum of 24°C during the tests. Further research to determine the significance of this parameter seems to be indicated. An important limitation of our study was the indirect measurement of blood flow by skin temperature. The use of other,
BLOOD FLOW RESPONSE AFTER COLD PROVOCATION, van Marum
more direct, techniques for assessing skin blood flow (eg, the laser Doppler technique) can give more detailed information on vasoactivity after an external stimulus and therefore should be used in future research. CONCLUSION The information provided by this study can help in the assessment of patients at risk for developing pressure ulcers. In clinical practice, pressure ulcer risk assessment is mainly performed by using risk scores. An important shortcoming of these scores is the fact that their validity is based on specific patient populations. Patient outcome on a risk score is always a reflection of the mean risk for all patients with the same score in that specific population. Direct measurement of the patient’s individual susceptibility therefore should be preferred. Further research should focus on developing techniques that can do this and that can be used in clinical practice. References 1. Meijer JH, Germs PH, Schneider H, Ribbe MW. Susceptibility to decubitus ulcer formation. Arch Phys Med Rehabil 1994;75:31823. 2. Agency for Health Care Policy and Research. Pressure ulcers in adults; prediction and prevention. Clinical Practice Guideline No. 3. Rockville (MD): US Department of Health and Human Services; May 1992. AHCPR Publ No. 92-0047. 3. Bergstrom N, Demuth PJ, Braden BJ. A clinical trial of the Braden scale for predicting pressure sore risk. Nurs Clin North Am 1987;22:417-28. 4. Norton D, McLaren R, Exton-Smith AN. An investigation of geriatric nursing problems in hospital. London: Churchill Livingstone; 1962. p 194-236. 5. Gosnell DJ. Pressure sore risk assessment: a critique. Part I. The Gosnell scale. Decubitus 1989;2:32-8. 6. Waterlow J. Calculating the risk. Nurs Times 1987;83:59-61. 7. Knoll Pharmaceutical Co. The Knoll scale of liability to pressure sores. In: McFarlane S, Castledine G, editors. St. Louis: Mosby; 1977.
43
8. van Marum RJ, Ooms ME, Ribbe MW, van Eijk J. Decubitus risk assessment using the CBO risk assessment scale for decubitus in a nursing home population. Age Ageing 2000;29:63-8. 9. van Marum RJ, Meijer JH, Bertelsmann FW, Ribbe MW. Impaired blood flow response following pressure load in diabetic patients with cardiac autonomic neuropathy. Arch Phys Med Rehabil 1997;78:1003-6. 10. Moorhouse JA, Carter SA, Doupe J. Vascular responses in diabetic peripheral neuropathy. Br Med J 1966;i:883-8. 11. Donk AF, Faes TJ, Broere D, van der Veen EA, Bertelsman FW. Quantitation of skin vasomotor control in normal subjects and in diabetic patients with autonomic neuropathy. J Neurol 1990;237: 457-60. 12. Faes TJ, Wagemans MF, Cillekens JM, Scheffer GJ, Karemaker JM, Bertelsmann FW. The validity and reproducibility of the skin vasomotor test—studies in normal subjects, after spinal anaesthesia, and in diabetes mellitus. Clin Auton Res 1993;3:319-24. 13. Faes TJ, Yff GA, De Weerdt O, Lanting P, Heimans JJ, Bertelsmann FW. Treatment of diabetic autonomic neuropathy with an aldose reductase inhibitor. J Neurol 1993;240:156-60. 14. Clifford C, Martin MF, Sheddon EJ, Kirby JD, Baird RN, Dieppe PA. Treatment of vasospastic disease with prostaglandin E. Br Med J 1980;281:1031-4. 15. Hansteen V. Medical treatment in Raynaud’s disease. Acta Chir Scand Suppl 1976;465:87-91. 16. Luck P, Wakeling A. Increased cutaneous vasoreactivity to cold in anorexia nervosa. Clin Sci (Colch) 1981;61:559-67. 17. Cleophas TJ, Fennis JF, van’t Laar A. Finger temperature after a finger-cooling test; influence of air temperature and smoking. J Appl Physiol 1982;52:1167-71. 18. Meijer JH, Schut GL, Ribbe MW, et al. Method for the measurement of susceptibility to decubitus ulcer formation. Med Biol Eng Comput 1989;27:502-6. 19. Mein CJ, Wagemans MF, Faes TJ, Meijer JH, Ribbe MW. Skin temperature response to a pressure load: studies in subjects before and during spinal anaesthesia. Arch Phys Med Rehabil 1995;76: 243-5. Supplier a. AD590; Analog Devices, One Technology Way, PO Box 9106, Norwood, MA 02062-9106.
Arch Phys Med Rehabil Vol 83, January 2002
The Relationship Between Pressure Ulcers and Skin Blood Flow Response After a Local Cold Provocation Rob J. van Marum, MD, PhD, Jan H. Meijer, PhD, Miel W. Ribbe, MD, PhD ABSTRACT. van Marum RJ, Meijer JH, Ribbe MW. The relationship between pressure ulcers and skin blood flow response after a local cold provocation. Arch Phys Med Rehabil 2002;83;40-3. Objective: To study the relationship between an impaired blood flow response after a local cold stimulus, testing nerve regulation of the local blood flow response, and an increased risk of developing pressure ulcers. Design: An observational, longitudinal, prospective study. Setting: Dutch nursing home. Patients: Eighty-two newly admitted somatic nursing home patients, age 60 years and older. Intervention: A local cold stimulus (17°C) applied to the trochanter major. Main Outcome Measures: On admission, blood flow response to a local cold stimulus. As the stimulus was withdrawn, the temperature measured at the skin increased asymptotically toward the final temperature, Tf. The velocity of this rise was characterized by the time constant, , of the process. On admission, and weekly during a 4-week follow-up period, the presence or absence of pressure ulcers was verified. Results: The blood flow response time correlated significantly with the risk of developing pressure ulcers. The patients who developed pressure ulcers during the follow-up period had a significantly longer response time than the patients who did not. Conclusions: Malfunction of the nervous regulatory mechanisms of the local blood flow is partially responsible for an increased susceptibility to pressure ulcer formation. Key Words: Autonomic nervous system; Elderly; Pressure ulcers; Rehabilitation; Skin temperature. © 2002 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation ARLY RECOGNITION OF PATIENTS at risk for developing pressure ulcers has always been an important cliniE cal problem. The risk of developing pressure ulcers results from a combination of 2 different sets of factors, the so-called external factors and internal factors.1,2 External factors are those related to the pressure (intensity, duration, direction); the internal factors (eg, fever, anemia, endothelial dysfunction, malnutrition) together determine a patient’s susceptibility to
From the Department of General Practice, Nursing Home Medicine and Social Medicine/Institute for Research in Extramural Medicine (van Marum, Ribbe); and the Department of Clinical Physics and Informatics (Meijer), Faculty of Medicine, Vrije Universiteit, Amsterdam, The Netherlands. Accepted in revised form January 11, 2001. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the author(s) or upon any organization with which the author(s) is/are associated. Reprint requests to Rob J. van Marum, MD, PhD, Dept of Geriatric Medicine, Bosch Medicentrum, PO Box 90153, ’s-Hertogenbosch, The Netherlands, e-mail: [email protected]. 0003-9993/02/8301-6480$35.00/0 doi:10.1053/apmr.2002.26827
Arch Phys Med Rehabil Vol 83, January 2002
developing pressure ulcers. This susceptibility is a patientbound characteristic that changes with time, depending on health status.1 Measuring this susceptibility has always been difficult because the underlying pathophysiologic mechanisms are unclear. Therefore, in most pressure ulcer risk assessment scores, a patient’s susceptibility to developing pressure ulcers is influenced by general physical condition, diabetes mellitus, nutritional status, and temperature.3-8 However, Meijer et al1 found that the recovery time of the local blood flow after relief of a local test pressure assessed the susceptibility to pressure ulcer formation. Susceptible patients were found to have substantially prolonged recovery times. Meijer1 calculated a patient’s risk of developing pressure ulcers by combining the outcome of the measurement of susceptibility with a measurement of the external forces acting on tissues (pressure, shear forces). The method used by Meijer,1 the so-called pressure-time-temperature (PTT) method, was based on an intense local pressure stimulus, which could be painful or induce local damage to the tissues. The objective of our study was to investigate the relationship between impaired blood flow response and increased risk for developing pressure ulcers, without using an external pressure stimulus to provoke this blood flow response. Blood flow response after pressure relief is determined by various individual mechanisms. In local autoregulation, vasoactive chemical agents formed during ischemia play an important role.1 The elastic mechanical properties of the tissue must be of importance because these properties at least determine the penetration depth of an exercised pressure, and thus the severeness of damage resulting from ischemia or distortion of the local endothelium and other tissues. The method used by Meijer,1 based on pressure as an external stimulus for the cessation of local blood flow, mainly tests local autoregulatory mechanisms. However, blood flow is also regulated by the autonomic nervous system. In a study9 of diabetic patients with autonomic neuropathy, it was suggested that impairment of the local blood flow response was the result of both endothelial damage and autonomic neuropathy. Because the main function of the autonomic vasoregulation of the skin is thermoregulation, applying a temperature stimulus can test the autonomic nervous system. The vascular response after a low-temperature stimulus is not dependent on muscle tone and mechanical deformation of the tissue, but solely on sympathetic regulation. The use of a low-temperature stimulus to provoke a local blood flow response is neither painful nor harmful to the patient. Our hypothesis was that the risk for developing pressure ulcers can be assessed by measuring the velocity of the blood flow response after a local cold provocation. METHODS Design An observational, longitudinal, prospective design was used to study the relationship between the blood flow response after a local cold stimulus and the susceptibility to the development of pressure ulcers in a cohort of newly admitted, elderly nurs-
BLOOD FLOW RESPONSE AFTER COLD PROVOCATION, van Marum
41
ing home patients. Blood flow response after a local cold stimulus was measured on admission, and the patients were examined for the existence or occurrence of pressure ulcers. During a 4-week follow-up period after admission, the patients were examined once a week for the development of new pressure ulcers. The protocol was approved by the ethics review board of the Academic Hospital of the Vrije Universiteit in Amsterdam. Subjects Data were gathered during a 34-week period from a cohort of newly admitted patients (N ⫽ 145), in a nursing home in Amsterdam, who met the following criteria: aged 60 years or older and admitted for somatic care. After application of the selection criteria, informed consent was obtained from each patient by the nursing home physician, who also assessed the patient’s suitability for participation on the basis of medical and ethical considerations (patients had to be able to maintain a horizontal position, lying on their side for at least 30min). If measurements could not be taken within 48 hours after admission, the patient was excluded. During the study period, patients could also be excluded for the following reasons: a substantial change in health status (these changes were considered to result in a health status that was no longer comparable with that on admission), discharge or death within 2 weeks after admission without developing pressure ulcers (participation in the study was then considered to be too short to observe the occurrence of pressure ulcers at the different intervals), and a pressure ulcer on admission. It was not possible to allocate a patient with pressure ulcers on admission to the group with newly developed pressure ulcers or to the group with no newly developed ulcers because stabilization or cure of the ulcers may well have been the result of extra efforts of the (para-) medical and nursing staff, which is difficult to quantify. Measurement of Blood Flow Response One task of the sympathetic nerve system is to regulate peripheral blood flow in response to temperature loads. Because damage of (peripheral) autonomic nerve fibers occurs in several diseases, widespread use has been made of temperature provocation tests to detect and assess the extent of this damage, and also to evaluate the effect of therapies on, for instance, diabetes mellitus, Raynaud’s disease, and anorexia nervosa.10-16 In these tests, local temperature stimuli (mainly cold) are applied locally, and local vascular responses are studied, usually by monitoring the local temperature course. The stimuli are usually applied to an extremity. In our study, a modified version of the method described by Hansteen15 and Cleophas et al17 was used, that is, the stimulus was not applied to the fingers but to the region of the greater trochanter. This site was chosen because it has been found to be a site at major risk for the development of pressure ulcers and that has been studied frequently. The regulatory functioning of the local sympathetic nerve system was tested by applying a local, low-temperature stimulus. The low-temperature stimulus was applied to the trochanter major region. Because patients had to keep their position for at least 30 minutes, they could choose the measurement side (left or right trochanteric region). The stimulus was applied for 7 minutes, by means of a cylindric probe with a contact surface area of 5.6cm2, through which water at a temperature of 17°C was pumped. The temperature measurements and signal processing were performed in a manner that has been described.18 The only difference in this study was that the temperature sensora was not potted into an indenter, but directly attached to
Fig 1. Example of registration representing a slow response after relief of the low-temperature stimulus. The cold stimulus is applied at T0 and removed at T1. The discrete steps in the temperature curves originate from the analog-to-digital conversion. Each step is 0.1°C.
the skin with adhesive tape. The measurements were performed in a temperature-controlled room. While lying on their side, patients were covered with a blanket, leaving 1 trochanter major exposed. They were then allowed to stabilize in the measurement position and to adapt to the environment during a period of 15 minutes before measurement. A physicist, who was unaware of the medical status of the patient, performed the measurements. Corresponding with the PTT measurements by Meijer et al,1 in which pressure was used as an external stimulus for provocation of the blood flow response, the method using temperature as a stimulus can be named the temperature-temperature-time (TTT) method. Figure 1 shows a typical example of registration. The initial skin temperature is T0. As the stimulus is applied, the skin temperature drops toward a new equilibrium (descending part of the curve). This process is dominated by the stimulus, and was not taken into account in the analysis. As the stimulus was withdrawn, the temperature measured at the skin increased asymptotically toward the final temperature, Tf. The velocity of this rise can be characterized by the time constant, , of the process ( ⬇ 1.44 times the half-life, t1/2, of the process). Both and Tf are calculated by means of exponential regression analysis, in which an exponential function is fitted to the ascending part of the curve. Although the amplitude of the response is considered to be dominated by the amplitude of the stimulus, and to contain no clinically relevant information, the difference in final temperature and initial temperature (Tf ⫺ T0) may be clinically relevant. In this study, both and Tf ⫺ T0 were included as independent variables. Measurement of Health Condition Health condition was operationalized according to the Norton score.4 A change in Norton score of more than 3 points during the follow-up period was considered to indicate a substantial change in health condition. The Norton score of each patient was routinely measured by the research physician on admission and after 4 weeks, or if necessary more frequently during the study period. Measurement of Pressure Ulcers The examinations to determine the existence of pressure ulcers during the study period were performed by a research Arch Phys Med Rehabil Vol 83, January 2002
42
BLOOD FLOW RESPONSE AFTER COLD PROVOCATION, van Marum
Table 1: Mean Response Time and Difference Between Final and Initial Skin Temperature for Patients With and Without Pressure Ulcers After 4 Weeks Patient Group
(min)
Tf ⫺ T0 (°C)
Group D⫹ (n ⫽ 21) Group D⫺ (n ⫽ 61) P
4.5 ⫾ 1.2 3.7 ⫾ 0.7 ⬍.0005
⫺1.6 ⫾ 0.8 ⫺1.4 ⫾ 0.6 NS
NOTE: Values are mean ⫾ SD. Abbreviation: NS, not significant.
physician who was unaware of the results of the measurement of the blood flow response. The first examination was performed within 48 hours after admission to the nursing home. Each patient was examined from head to toe for the presence or absence of pressure ulcers. A pressure ulcer was defined as any degenerative change, caused under influence of pressure and shear forces acting on biologic tissues. In this study, nonblanching hyperemia was considered to be the initial manifestation of a pressure ulcer. Analysis The mean response time () and the difference between final and initial temperature measurement (Tf ⫺ T0) of the group of patients with newly developed pressure ulcers (ulcers developed during follow-up period regardless of location) were compared with those of the group of patients with no newly developed pressure ulcers by means of Student’s t test. The relationship between the time difference in skin temperature, the response-time, and the risk of developing pressure ulcers was calculated by means of logistic regression analysis. These relationships were adjusted for age, gender, and prevention given. RESULTS Of the 145 patients who met the inclusion criteria, 29 patients were not measured for the following reasons: refusal (n ⫽ 6), assessed as unsuitable by the nursing home physician for medical or ethical reasons (n ⫽ 13), or measurement not performed within 48 hours after admission (n ⫽ 10). Of the remaining 116 patients who were measured, 34 patients (29%) were excluded for the following reasons: the presence of a pressure ulcer on admission that remained stable or was cured during the follow-up period (n ⫽ 20), discharge or death within 2 weeks after admission without developing a pressure ulcer (n ⫽ 9), and substantial change in health condition (n ⫽ 5). Data on 82 patients were included in the analyses. In total, 21 patients (26%) developed a pressure ulcer during the study period. Blood flow responses of patients in the group who developed pressure ulcers (D⫹ group) were compared with the responses of those who did not, develop pressure ulcers (D⫺ group) (table 1). D⫹ group patients showed a significantly higher time constant than D⫺ group patients. No differences between the groups could be found with regard to the difference between final and initial skin temperature. The relationship between (range, 130 – 436s; mean ⫾ standard deviation [SD], 235 ⫾ 53s) and the risk (R) of developing pressure ulcers was significant (P ⫽ .006). This relationship showed no significant change when prevention was included as a confounder, and can be described with the formula: y ⫽ ⫺8.23 ⫹ .0159, in which y ⫽ loge(Pulcer /1 ⫺ Pulcer), where Pulcer ⫽ pressure ulcer. For each value of y, R can be calculated: R ⫽ ey (1 ⫹ ey). Arch Phys Med Rehabil Vol 83, January 2002
No significant relationship was found between Tf ⫺ T0 and the risk of developing pressure ulcers. DISCUSSION The objective of this study was to investigate whether a relationship exists between the blood flow response after a local cold stimulus, using the TTT method, and the risk for developing pressure ulcers. In this prospective evaluation, a significant relationship was found between (the response time of the skin temperature after relief of the low-temperature stimulus) and the risk for developing pressure ulcers. The D⫹ group showed a significantly slower response time compared with the D⫺ group (table 1). From these observations, it can be concluded that malfunction of the nervous regulatory mechanisms of the local blood flow is at least partially related to increased susceptibility to pressure ulcer formation. These findings support the observations reported by Mein et al,19 who investigated the influence of spinal anesthesia on the local blood flow response as a model for autonomic neuropathy. They reported a doubling of the blood flow response time after relief of a pressure stimulus during spinal anesthesia. However, in their study, no distinction could be made between the influence of blockage of the autonomic nerves and the motor nerves, resulting in paralysis of muscle tissue. Moreover, the stimulus used in their study (local pressure) induced tissue strain.19 Because in our study no tissue strain was induced, it can be concluded that the delayed blood flow response, which was also reported by Mein,19 has to be attributed to malfunction of the autonomic nervous system. The studies performed by Meijer1 and Mein,19 based on a pressure stimulus, show a much greater variation in the blood flow response, compared with the relatively small variation observed in the response provoked by a temperature stimulus, whereas both responses correlated with the risk of developing pressure ulcers. This may imply that both responses are related, at least partially, to different sets of subfactors of susceptibility. However, the reliability of the 2 methods may differ, so this assumption should be made with caution. It cannot be excluded that the experimental setting influenced the variations. There is an important difference between the pressure stimulus test and the temperature stimulus test. With the pressure stimulus, at the end of the stimulation period there is thermal equilibrium in the tissue, whereas with the temperature stimulus, there is an absence of thermal equilibrium. This implies that the velocity of the temperature response after relief of the temperature stimulus is modulated by the thermal diffusion flux, which complicates comparison of the responses of the 2 methods. The thermal diffusion flux may well be influenced by the amount of subcutaneous fat tissue. A substantial variation in the difference between final and initial skin temperature (Tf ⫺ T0) was observed. The skin temperature did not return to its original value after relief of the temperature stimulus in either group of patients. These observations suggest a persisting vasoconstriction. However, no relationship with the risk for developing pressure ulcers could be established. A similar, substantial variation was found in a previous study10 based on a comparable cold provocation test, in which fingers were cooled. However, the reported possible influence of ambient temperature on Tf ⫺ T0 does not apply to the results of our study because room temperature was kept at a minimum of 24°C during the tests. Further research to determine the significance of this parameter seems to be indicated. An important limitation of our study was the indirect measurement of blood flow by skin temperature. The use of other,
BLOOD FLOW RESPONSE AFTER COLD PROVOCATION, van Marum
more direct, techniques for assessing skin blood flow (eg, the laser Doppler technique) can give more detailed information on vasoactivity after an external stimulus and therefore should be used in future research. CONCLUSION The information provided by this study can help in the assessment of patients at risk for developing pressure ulcers. In clinical practice, pressure ulcer risk assessment is mainly performed by using risk scores. An important shortcoming of these scores is the fact that their validity is based on specific patient populations. Patient outcome on a risk score is always a reflection of the mean risk for all patients with the same score in that specific population. Direct measurement of the patient’s individual susceptibility therefore should be preferred. Further research should focus on developing techniques that can do this and that can be used in clinical practice. References 1. Meijer JH, Germs PH, Schneider H, Ribbe MW. Susceptibility to decubitus ulcer formation. Arch Phys Med Rehabil 1994;75:31823. 2. Agency for Health Care Policy and Research. Pressure ulcers in adults; prediction and prevention. Clinical Practice Guideline No. 3. Rockville (MD): US Department of Health and Human Services; May 1992. AHCPR Publ No. 92-0047. 3. Bergstrom N, Demuth PJ, Braden BJ. A clinical trial of the Braden scale for predicting pressure sore risk. Nurs Clin North Am 1987;22:417-28. 4. Norton D, McLaren R, Exton-Smith AN. An investigation of geriatric nursing problems in hospital. London: Churchill Livingstone; 1962. p 194-236. 5. Gosnell DJ. Pressure sore risk assessment: a critique. Part I. The Gosnell scale. Decubitus 1989;2:32-8. 6. Waterlow J. Calculating the risk. Nurs Times 1987;83:59-61. 7. Knoll Pharmaceutical Co. The Knoll scale of liability to pressure sores. In: McFarlane S, Castledine G, editors. St. Louis: Mosby; 1977.
43
8. van Marum RJ, Ooms ME, Ribbe MW, van Eijk J. Decubitus risk assessment using the CBO risk assessment scale for decubitus in a nursing home population. Age Ageing 2000;29:63-8. 9. van Marum RJ, Meijer JH, Bertelsmann FW, Ribbe MW. Impaired blood flow response following pressure load in diabetic patients with cardiac autonomic neuropathy. Arch Phys Med Rehabil 1997;78:1003-6. 10. Moorhouse JA, Carter SA, Doupe J. Vascular responses in diabetic peripheral neuropathy. Br Med J 1966;i:883-8. 11. Donk AF, Faes TJ, Broere D, van der Veen EA, Bertelsman FW. Quantitation of skin vasomotor control in normal subjects and in diabetic patients with autonomic neuropathy. J Neurol 1990;237: 457-60. 12. Faes TJ, Wagemans MF, Cillekens JM, Scheffer GJ, Karemaker JM, Bertelsmann FW. The validity and reproducibility of the skin vasomotor test—studies in normal subjects, after spinal anaesthesia, and in diabetes mellitus. Clin Auton Res 1993;3:319-24. 13. Faes TJ, Yff GA, De Weerdt O, Lanting P, Heimans JJ, Bertelsmann FW. Treatment of diabetic autonomic neuropathy with an aldose reductase inhibitor. J Neurol 1993;240:156-60. 14. Clifford C, Martin MF, Sheddon EJ, Kirby JD, Baird RN, Dieppe PA. Treatment of vasospastic disease with prostaglandin E. Br Med J 1980;281:1031-4. 15. Hansteen V. Medical treatment in Raynaud’s disease. Acta Chir Scand Suppl 1976;465:87-91. 16. Luck P, Wakeling A. Increased cutaneous vasoreactivity to cold in anorexia nervosa. Clin Sci (Colch) 1981;61:559-67. 17. Cleophas TJ, Fennis JF, van’t Laar A. Finger temperature after a finger-cooling test; influence of air temperature and smoking. J Appl Physiol 1982;52:1167-71. 18. Meijer JH, Schut GL, Ribbe MW, et al. Method for the measurement of susceptibility to decubitus ulcer formation. Med Biol Eng Comput 1989;27:502-6. 19. Mein CJ, Wagemans MF, Faes TJ, Meijer JH, Ribbe MW. Skin temperature response to a pressure load: studies in subjects before and during spinal anaesthesia. Arch Phys Med Rehabil 1995;76: 243-5. Supplier a. AD590; Analog Devices, One Technology Way, PO Box 9106, Norwood, MA 02062-9106.
Arch Phys Med Rehabil Vol 83, January 2002