MEASUREMENT OF TRANSCUTANEOUS OXYGEN PRESSURE IN NORMAL AND ISCHAEMIC SKIN

Transcription

1 MEAUREMENT OF TRANCUTANEOU OXYGEN PREURE IN NORMAL AND ICHAEMIC KIN G.. E. DOWD, K. LINGE, G. BENTLEY From the University Department oforthopaedic and Accident urgery, Royal Liverpool Hospital The transcutaneous oxygen pressure (tcpo2) was measured by a polarographic technique in the legs of 161 volunteers and compared with the levels found in 62 patients with ischaemic skin due to peripheral vascular disease. The results show that the tcpo2 was related to the degree of ischaemia and, in many cases, was a more accurate guide to the viability of the skin than clinical assessment. Measurement of the transcutaneous oxygen pressure in the leg at the site of amputation in 24 patients with peripheral vascular disease showed that a preoperative level greater than 40 millimetres of mercury at an electrode temperature of 44 degrees Celsius was necessary for the skin of the stump to heal. The technique is simple, non-invasive and reliable. The tcpo2 accurately reflects the physiological and pathological changes in the circulation of the skin. It has potential in many fields of surgery where careful assessment of the viability of the skin is necessary. An accurate method of assessing the viability of skin would be invaluable in many ischaemic conditions including peripheral vascular disease and limb trauma associated with cutaneous damage and vascular injury. At present, no satisfactory method is available which is non-invasive, accurate and simple to use. Previous methods of investigating the viability of the skin have included the measurement of cutaneous blood pressure using a photo-electric probe (Nielsen, Poulsen and Gyntelberg 1973), the measurement of cutaneous blood flow by a thermal conductance method (Challoner 1975), intradermal measurement ofoxygen tension (pence and Walker 1976), and more recently the application of radioisotope tracers to measure blood flow and distal blood pressure (Kostuik et al. 1976; Holstein et al. 1978). All these techniques have been applied to clinical situations but have limitations in their routine use ; several methods are invasive and the others are technically difficult to use or cause the patient discomfort. Recently, a transcutaneous oxygen monitor has been developed as a non-invasive method of continuous monitoring ofthe arterial oxygen pressure (Pa,o2) (Huch, Lubbers and Huch 1972). It consists of a Clark-type electrode which is attached to the surface of the skin by a self-adhesive ring. Oxygen diffusing through the skin is G.. E. Dowd, MChOrth, FRC, enior Lecturer and Honorary Consultant K. Linge, Research Assistant University Department of Orthopaedic and Accident urgery, Royal Liverpool Hospital, Prescot treet, Liverpool L7 8XP, England. Professor G. Bentley, ChM, FRC Royal National Orthopaedic Hospital, 234 Great Portland treet, London WiN 6AD, England. Requests for reprints should be sent to Mr G.. E. Dowd British Editorial ociety of Bone and Joint urgery X/83/ $2.00 measured by a polarographic technique and the level is displayed on a monitor as the transcutaneous oxygen pressure (tcpo2). When the skin below the electrode is heated to produce maximal vasodilatation of the capillanes, the increase in blood flow results in an increase in oxygen diffusion through the skin. It has been shown that with maximal vasodilatation of the capillaries the Pa,o2 correlates closely with the tcpo2 in adults (Rooth et a!. 1976; Gothgen and Jacobsen 1978). The Pa,o2 and tcpo2 are physiologically different and the relationship between the two depends on many factors, in particular, the blood flow under the electrode. Eickhoff, Ishihara and Jacobsen (1979) have shown that the tcpo2 is directly related to blood flow in normal skin. They have also shown that the tcpo2 depends on the perfusion pressure of blood under the electrode. Gothgen and Jacobsen (1978) have similarly observed that the tcpo2 is directly related to the mean arterial blood pressure. In chronic ischaemic states there is a fall in cutaneous blood flow and perfusion pressure (Holstein and Lassen 1973 ; Kostuik et a!. 1976) and the oxygen pressure in the skin is reduced (pence and Walker 1976). The transcutaneous oxygen pressure should therefore reflect all these changes and the tcpo2 measurement should relate to the degree of ischaemia, thereby allowing an accurate assessment of the viability of the skin. In order to test this hypothesis, the tcpo2 was measured in a group of normal volunteers and the results compared with those in a series of patients with varying degrees of ischaemia resulting from peripheral vascular disease. The technique has subsequently been applied to patients undergoing amputation for peripheral vascular disease and to patients undergoing operations for orthopaedic conditions and trauma where the viability of VOL. 65-B, No. I, JANUARY

2 80 G.. E. DOWD, K. LINGE, G. BENTLEY the skin was in question. By measuring the tcpo2 near to the site of the incision before operation, it was hoped to identify the tcpo2 above which the skin might confidently be expected to heal. MATERIAL AND METHOD The transcutaneous oxygen pressure was measured using a Radiometer TCM1TC Oxygen Monitor connected to a chart recorder (Fig. 1). The electrode was calibrated to a zero level of oxygen in a bisulphite buffer solution and to atmospheric oxygen pressure. In order to standardise the measurements, the patients and volunteers were rested and asked to remain silent for 20 minutes before the application of the electrode and during measurement. Measurements were taken with the subjects lying on a bed with their legs horizontal to the floor. None of the patients had severe respiratory disease. The electrode was attached to the skin by a self-adhesive ring and the heater in the electrode was set to produce a temperature of44 degrees Celsius. The temperature of the surrounding air was 22 degrees Celsius ( ± one degree Celsius). The tcpo, recording stabilised approximately 20 minutes after application of the electrode C I,. n=161 mean= 70mm Hg D= ± TcPO2 (mm Hg) Fig. 2 Histogram of the tcpo2 recorded in the legs of 16 1 normal volunteers. REULT -\.. A - - Fig. The apparatus: A, electrode: B, monitor: C, calibration unit; and D, chart recorder. I Normal volunteers. Measurements of the tcpo2 on the dorsum of the foot in 161 normal volunteers showed a range of 45 to 95 millimetres of mercury with a mean of 70 millimetres of mercury, standard deviation ± nine millimetres of mercury (Fig. 2). When these measurements were related to age, the age-dependent deviations were not significant. A similar range of transcutaneous measurements was found on the chest wall and 10 centimetres above and below the knee (Table I). When measurements at various sites were compared in the same volunteer it was found that there was a closer correlation between areas of skin which were anatomically similar. In other words, measurements taken on the dorsum of both feet, or on the skin 10 centimetres below each knee were more closely related than a chest Normal volunteers. The tcpo, was measured on the skin of the dorsum ofthe foot in 161 normal volunteers, with an age range of 12 to 94 years (mean 45 years). Further measurements were made in 91 of these volunteers on the skin of the chest wall over the second rib interspace, in the left midclavicular line. In 25 volunteers, measurements were also taken on the skin at points 10 centimetres above and below the knee. Patients with peripheral vascular disease. ixty-two patients, with an age range of 44 to 85 years (mean 66 years), suffering from peripheral vascular disease had readings taken on the dorsum of the affected foot. The patients were divided into three groups depending on the degree of ischaemia of the skin assessed clinically : patients with symptoms of intermittent claudication without evidence of ischaemic skin ; patients with signs of ischaemic skin on the dorsum of the foot but without evidence of gangrene: and patients with gangrene of the skin of the foot (the tcpo, was measured on the skin adjacent to the gangrenous tissue because of the difficulty in maintaining adhesion of the electrode directly onto necrotic tissue). Patients undergoing amputation. In 24 patients with an age range of 50 to 85 years (mean 67 years) the tcpo2 was measured before amputation for gangrene due to peripheral vascular disease. Readings were taken on the dorsum of the foot and on the skin 10 centimetres above and below the knee near the sites of routine amputations. The decision was made by the surgeon regarding the site of amputation based on clinical examination without reference to the tcpo,. The amputation stumps were examined after operation and classified in relation to primary healing of the skin, delayed healing or failure to heal. Table I. Comparison of transcutaneous oxygen pressures at various sites in normal volunteers Number tcpo2 (mmhg) of ite of measurement volunteers Range Mean D Chest ±11 10 centimetres below ± 9 knee 10 centimetres below ± 9 knee Dorsum of foot ± 11 Left leg 10 centimetres ± 9 below-knee Right leg 10 centimetres ± 8 below-knee Dorsum ofthe left foot ± 12 Dorsum of the right foot ± 1 1 THE JOURNAL OF BONE AND JOINT URGERY

3 :.J MEAUREMENT OF TRANCUTANEOU OXYGEN PREURE IN NORMAL AND ICHAEMIC KIN 81 measurement compared with a below-knee or foot measurement. Presumably the reason for this was differences in thicknessofthe skin and other physiological factors affecting the skin on the chest and leg. Comparison of the measurements taken at the various sites in the legs and on the chest wall failed to show an oxygen pressure gradient from proximal to distal parts of the body in normal volunteers. Patients with peripheral vascular disease. The tcpo2 on the dorsum of the foot in patients with intermittent claudication but without skin changes ranged from 26 to 72 millimetres of mercury with a mean of 52 millimetres of mercury. Patients with signs of Ischaemic skin had a range of tcpo2 of 19 to 49 millimetres of mercury with a mean of 33 millimetres of mercury, whilst measurements adjacent to gangrenous tissue were in the range of 0 to 38 millimetres of mercury with a mean of 10 millimetres of mercury. When the histograms relating the percentage of patients in each group to the tcpo2 on the dorsum of the foot were compared to the histogram of the normal volunteers within the same age range, the abnormal groups clearly showed a shift to the left of normal (Fig. 3). The degree of shift was related to the severity of the ischaemia. Patients with symptoms of intermittent claudication but without signs of ischaemia had tcpo2 measurements encompassing the lower range of normal, with 24 per cent of patients below the lower limits of normal (45 millimetres of mercury). Ninety-three per I Normal (n73)..r rrrrrli., cent of the patients with clinical signs of ischaemic skin had a tcpo2 below the lower limits of normal ; the remaining seven per cent were within five millimetres of mercury of the lower normal limit. Measurements taken adjacent to gangrenous tissue approached zero millimetres of mercury in 52 per cent of patients, whilst all readings were below the lower limits of normal. It should be stressed that readings in the latter group were taken adjacent to gangrenous tissue on skin of varying degrees of viability. When individual patients were examined and the readings from the thigh, leg and foot were compared, they all showed a transcutaneous oxygen pressure gradient from the proximal to distal part of the limb. This gradient was not apparent in normal volunteers and was related to the degree of ischaemia. Table II. Transcutaneous oxygen pressure at the level of amputation related to subsequent healing of the stump tcpo2 at site of Type of amputation Case Age ex amputation (mmhg) Result 1 71 F Above-knee 74 Healed 2 74 M Above-knee 64 Healed 3 52 M Above-knee 63 Healed 4 65 M Above-knee 62 Healed 5 52 M Above-knee 57 Healed 6 80 F Above-knee 56 Healed 7 53 M Above-knee 55 Healed 8 61 M Midtarsal 54 Healed 40 I, Claudicatlon ( n 15) G nr m 9 71 M Below-knee 53 Healed M Below-knee 49 Delayed healing M Below-knee 48 Healed M Above-knee 48 Healed M Above-knee 48 Healed M Above-knee 48 Healed M Above-knee 45 Healed M Above-knee 45 Healed Oangr.n. (.=33) M Above-knee 41 Healed M Above-knee 40 Healed M Midtarsal 40 Failed C : 20 I F Toe 36* Failed F Midtarsal 36 Failed TcPO2 Fig. 3 (mmhg) Histograms of the four groups of volunteers and patients, relating the percentage of individuals in each group to the tcpo2 measurement on the dorsum of the foot F Below-knee 35 Failed M Below-knee 34 Failed M Toe 0 Failed * tcpo, on the dorsum of the foot VOL. 65-B, No. 1, JANUARY 1983

4 82 G.. E. DOWD, K. LINGE, 0. BENTLEY Amputees. Twenty-four patients with peripheral vascular disease of the lower limb judged on clinical grounds to require amputation for gangrene had the tcpo2 measured on the dorsum of the foot and on the skin 10 centimetres above and below the knee. Fourteen patients subsequently underwent above-knee amputation : five belowknee, three midtarsal and two had local amputation of a gangrenous toe (Table II). All above-knee amputations healed without difficulty and all had had a preoperative tcpo2 reading of 40 millimetres of mercury or above. Two of the below-knee amputations healed primarily with a further one healing after a delay of several weeks. All three had a tcpo2 of 48 millimetres of mercury or more 10 centimetres above the knee. The two below-knee amputations which failed to heal and required a more proximal amputation had tcpo2 readings of 34 and 35 millimetres of mercury before operation. Two of the three midtarsal amputations with a tcpo2 on the dorsum of the foot of 36 and 40 millimetres of mercury failed to heal. The midtarsal amputation which healed had a preoperative reading taken on the dorsum of the foot of 54 millimetres of mercury. Two patients underwent amputation of a gangrenous toe. They had readings of 0 and 36 millimetres of mercury on the dorsum of the foot, despite the fact that clinically the skin appeared to be well vascularised. Both failed to heal and a more proximal amputation was necessary. When the tcpo2 at the site of amputation was correlated with skin healing, all amputations performed in an area in which the tcpo2 was less than 40 millimetres of mercury failed to heal irrespective of the site of amputation (Fig. 4). Providing the tcpo2 was above 40 millimetres of mercury the amputation stump healed. In two patients the tcpo2 at the level of amputation (one midtarsal and the other above the knee) was 40 millimetres of mercury; the midtarsal stump failed to heal and the above-knee stump healed. DICUION Assessment of the viability of the skin is usually made on clinical grounds, taking into account such factors as colour, capillary return, alteration in temperature and presence or absence ofperipheral pulses. There are many situations, however, where clinical examination of the skin is inadequate and where a more objective measurement would be invaluable. Arteriography will demonstrate the patency of large or medium-sized vessels but will not provide information on the local cutaneous blood supply. The use of Doppler ultrasound has little value in investigating the small calibre, distal segments of the vascular tree (Harris et a!. 1974). Radioactive isotope tracers have been used to measure the pressure of blood perfusion in the cutaneous vessels (Holstein et a!. 1978), but the method requires special expertise and, by its invasive nature, is contraindicated in ischaemic states. Kostuik et a!. (1976) measured blood flow in the skin by measuring the rate of >, 4 C., C 5, a. 5, la 0Inn1 LI1 primary healing secondary healing LII failed to heal Tc P02 ( mm Hg) at site of amputation Fig. 4 Histogram relating the tcpo2 at the site ofamputation measured before operation to subsequent healing of the skin. clearance of xenon-133 gas from the tissue. They stated that a blood flow of less than 1.5 millilitres per minute per 100 grams of tissue would result in failure of the skin to heal after amputation ; however, in our experience consistency of results is a problem with this method. Theoretical considerations suggested that the measurement of the tcpo2 was related to changes in the blood circulation ofthe skin and this has been confirmed by the present study. A series of normal values has been established which can be compared with values found in ischaemic states. Values obtained for normal volunteers in this study show that the tcpo2, for practical purposes, is not affected by age. Moreover, there was no observable pressure gradient from the proximal to distal end of the limb in normal individuals. The measurements ofthe tcpo2 in groups of patients with increasingly severe peripheral ischaemia of the lower limbs show that they accurately reflect the clinical state of the skin. In patients with symptoms of intermittent claudication without clinically observable changes of the skin, the transcutaneous oxygen level was lower than normal in 24 per cent, presumably due to a decrease in the local circulation which was not apparent on clinical examination. When the preoperative oxygen pressure at the site of amputation for peripheral vascular disease was compared to skin healing, the skin of the stump healed providing that the tcpo2 was above 40 millimetres of mercury; pressure below 40 millimetres of mercury resulted in breakdown of the wound and failure to heal. A tcpo2 of 45 millimetres of mercury was the lower limit of normal volunteers. Thus it appears that a tcpo2 below 40 millimetres of mercury is always related to a severe circulatory disturbance in the cutaneous tissues. In patients with peripheral vascular disease the site of amputation is usually a compromise between the desire to amputate at the most distal level and the expectation of the stump to heal; it is well known that failure of the stump to heal is more likely with distal 75 THE JOURNAL OF BONE AND JOINT URGERY

5 MEAUREMENT OF TRANCUTANEOU OXYGEN PREURE IN NORMAL AND ICHAEMIC KIN 83 amputations. Review of the transcutaneous oxygen pressures in this series of patients where the site of amputation was decided on clinical grounds alone, shows that the higher transcutaneous oxygen pressures were associated with above-knee amputations and lower levels with below-knee and more distal amputations. It should be stressed that the levels of tcpo2 must always be related to the temperature of the electrode. In clinical practice, the temperature should be set at a level at which maximal vasodilatation ofthe cutaneous vessels will occur, but which will not damage the skin, thus allowing continuous monitoring. By experimentation it was found that 44 degrees Celsius was the most suitable temperature for the electrode. Wound healing is associated with many factors and is not solely dependent on blood flow. It is known that circulation of blood to the edges of the skin is a primary consideration. This may be deleteriously affected by infection and wound haematoma. The technique of measuring transcutaneous oxygen pressure has been used to a limited extent on skin before operations other than amputation. Patients admitted for a Keller s operation on the great toe with skin of dubious viability have been measured. One patient with a tcpo2 of 38 millimetres of mercury has failed to heal. Measurements have also been taken at the site of the skin incision before plating oftibial fractures 10 to 14 days after injury. o far, wound breakdown has only occurred in patients with a preoperative oxygen pressure of below 40 millimetres of mercury. Further studies on traumatised skin are necessary however, before such measurements can be realistically evaluated. We wish to thank our colleagues for permission to use their patients in this study and, in particular, Mr G. A. McLoughlin, Consultant Vascular urgeon, Royal Liverpool Hospital and Miss J. Doyle, who prepared the manuscript. REFERENCE Challoner AVJ. Accurate measurement of skin blood flow by a thermal conductance method. Med Biol Eng Comput 1975 ; 13 : Eickhoff J, Ishihara 5, Jacobsen. PaO, by skin electrode. Lancet 1979;ii : G#{246}thgenI, Jacobsen E. Transcutaneous oxygen tension measurement. I : age variation and reproducibility. Acta Anaesthesiol cand 1978 : uppl 67: Hams PL, Taylor LA, Cave FD, Charlesworth D. The relationship between Doppler ultrasound assessment and angiography in occlusive arterial disease of the lower limbs. urg Gynecol Obstet 1974; 138 : Holstein P, Lassen NA. Radioisotope clearance technique for measurement of distal blood pressure in skin and muscles. cand J C/in Lab Invest 1973:31, uppl 128: Holstein P, Noer 1, Tonnesen KH, ager P, Lassen NA. Distal blood pressure in severe arterial insufficiency in gangrene and severe ischaemia of the lower extremities. In : Bergan JJ, Yao JT, eds. Operative techniques in vascular surgery. New York : Grune and tratton 1978 ; Huch R, Lubbers DW, Huch A. Quantitative continuous measurement of partial oxygen pressure on the skin of adults and new-born babies. Pfluegers Arch 1972:337 : Kostuik JP, Wood D, Hornby R, Feingold, Mathews V. The measurement of skin blood flow in peripheral vascular disease by epicutaneous application of Xenon. J Bone Joint urg [Am] 1976;58-A : Nielsen PE, Poulsen HL, Gyntelberg F. kin blood pressure measured by a photoelectric probe and external counterpressure. cand J C/in Lab Invest 1973;31, uppl 128: Rooth G, Hedstrand U, Tyd#{233}nH, Ogren C. The validity of the transcutaneous oxygen tension method in adults. Crit Care Med 1976;4: pence VA, Walker WF. Measurement ofoxygen tension in human skin. Med Biol Eng Comput 1976; 14: VOL. 65-B, No. 1, JANUARY 1983