Assessment of peripheral vascular disease by postocclusive transcutaneous oxygen recovery time

Transcription

1 Assessment of peripheral vascular disease by postocclusive transcutaneous oxygen recovery time Harry B. Kram, M.D., Paul L. Appel, M.P.A., Rodney A. White, M.D., and William C. Shoemaker, M.D., Torrance and Los Angeles, Calif. A method for assessing peripheral vascular disease (PVD) was developed from the pattern of transcutaneous oxygen (Ptco2) changes after temporary limb ischemia induced by pneumatic blood pressure cuff occlusion. The transcutaneous oxygen recovery half-time (TORT) was defined as the time required to recover half of the decrease in the limb/chest Ptco2 ratio produced by temporary limb ischemia. TORT was examined in subjects with and without significant PVD. Patients who underwent operative therapy for symptomatic PVD were studied before and after operation. Comparison was also made of the values of patients in whom therapy was successful in resolving symptoms vs. those in whom it was not. TORT was found to improve the diagnostic accuracy of Ptco2 measurements to differentiate normal vs. PVD limbs and successfial vs. unsuccessful results of therapy. Normal subjects uniformly had TORT values ~1.5 minutes at both calf and foot positions; limbs with symptomatic PVD consistently had TORT values that were significantly longer in duration (p < 0.001). Limbs with PVD that received successfial therapy had a significant improvement in calf and foot TORT values after surgery (p < 0.01); postoperative values were usually slightly longer than those of normal subjects, probably reflecting some residual disease. Postoperatively, limbs with PVD that had received unsuccessful therapy either had only slight improvement or worsening of their TORT values, which were significantly greater than the TORT values of limbs that received successful therapy (p < 0.001). We conclude that measurement ofpostocclusive TORT is a reliable method for diagnosing PVD and for quantitatively evaluating residual disease after operative therapy. Furthermore, since this is a noninvasive, easily performed measurement that does not require patient effort, it is a practical quantitative method for repeated assessment of PVD severity. (J VASC SURG 1984; 1: ) Noninvasive transcutaneous 02 (Ptco2) monitoring is becoming increasingly useful in the detection and functional quantitation of arterial compromise in peripheral vascular disease (PVD). 1,2 Initial investigations demonstrated that much of the overlap in Ptco2 values between normal and diseased limbs could be attributed to variations in arterial oxygen content, cardiac output, and systemic oxygen delivery. However, use of the limb/chest Ptco2 ratio has helped to quantify the deficit of tissue perfusion independent of variations in total body circulation and oxygen delivery. 1,2 Previous investigators also have employed various tools to quantitate postocclusive From the Department of Surgery, Los Angeles County Harbor- UCLA Medical Center, Torrance, and the UCLA School of Medicine, Los Angeles. Reprint requests: William C. Shoemaker, M.D., Department of Surgery, Los Angeles County Harbor- UCLA Medical Center, Torrance, CA circulatory recovery, including measurement of reactive hyperemia and toe pulse reappearance time. Measurement of resting limb Ptco2 tension is a simple method to quantify local tissue perfusion as a means to evaluate arterial obstruction. 2"3 However, these values do not adequately reflect the functional capacity of the limb's response to stress. The accuracy of tests for PVD may be enhanced by stressing limb circulation: (1) to exaggerate the reduction in limb perfusion secondary to stenotic lesions and (2) to quantitate the tissue oxygenation response to this stress. Such "stress testing" is particularly important in patients in whom resting Ptco2 values do not establish the diagnosis of PVD. Previously, the regional Ptco2 response to treadmill exercise testing was reported. 1 This treadmill technique provides important physiologic information but may be unsuited for routine clinical screening, particularly in patients with rest pain, ischemic ulceration, ampu- 628

2 Volume 1 Number 5 September 1984 Assessment of PVD by postocdusive TORT 629 torr CHEST Ptc02 7O 60 /~- /1 : : _~ ~ - -" - I oeripherolvosculor I disease.50 ~.~ 7_ successful therapy unsuccessful therapy Conlro, Arl'eriol i 2 ~) 4-,5 Occlusion minutes offer arterial occlusion Fig. 1. Effect on chest Ptco2 of temporary limb ischemia for 4 minutes by standard pneumatic blood pressure cuff: in normal volunteer subjects, in patients with significant PVD, in patients who had successful therapy, and in patients who had unsuccessful therapy. Values are mean _+ SEM. Note that chest Ptco2 consistently increased during temporary limb ischemia and then remained above preocclusion levels throughout remainder of examination perical, this effect was most pronounced in normal subjects (p < 0.05 by paired t test). % IOO %0 I normal~.. successful /'~'/ \~therapy N~/~ / i, CALF i/~t " - T J. "~! unsuccessfu 'herapy " peripheral vascu,d ar isease FOOT tation, general debility, and cardiopulmonary disease. Moreover, when subjected to treadmill testing, patients with associated coronary atherosclerosis have a low but definite risk of myocardial infarction. 4 In addition, the muscular contractions that occur during exercise may hinder blood flow and thus nullify the effect of the increasing limb oxygenation. The extent of this hindrance is considerable in the calf, where maximal muscular contraction may actually arrest flow2 In this article we describe a simple technique to evaluate limb perfusion by measurement of regional Ptco2 tension after temporary limb ischemia. This noninvasive test measures postocclusive transcutaneous oxygen recovery half-time (TORT); its application in the diagnosis, evaluation, and perioperative management of symptomatic PVD is described. MATERIAL AND METHODS Subject selection. Paired measurements of calf and foot postocclusive TORT were performed 55 times in 41 limbs of 22 subjects. Of these, 11 patients were consecutive readmissions for operative therapy of symptomatic PVD. All symptomatic patients had interrnittent claudication or rest pain as a chief complaint. The remaining subjects in the study were normal volunteers. Group 1 comprised 22 limbs studied in healthy, normal volunteer subjects between 20 and 70 years of age; the mean age was (SD) years; these control subjects had no known lower extremity vascular disease. Group 2 o Control Arterial Occlusion i i i a g minutes after arterial occlusion Fig. 2. Effect of temporary arterial occlusion on calf and foot Ptc02 values, cxpressed as percentage of simultaneously recorded chest Ptc02: in normal volunteer subjects, in patients with significant PVD, in patients who had successful therapy, and in patients who had unsuccessful therapy. Values are mean-+ SEM. Note that postocclusive recovery is considerably delayed in limbs with symptomatic PVD. comprised 19 symptomatic limbs in 11 patients who had angiographic evidence of significant PVD. Thcse patients ranged in age from 21 to 63 years (mean years). Six patients were men; five were women; seven were heavy smokers and four had diabetes; one patient had prior vascular surgery and presented with intermittent claudication resulting from a thrombosed iliofemoral bypass graft. Group 3a comprised 12 of the 19 limbs of group 2 patients who were restudied 1 to 2 weeks after having undergone successful therapy; these included seven with aortofemoral bypass, four with femoralpopliteal bypass, and one with axillofcmoral bypass. Surgcry was successful in resolving symptoms in all group 3a patients. Group 3b comprised 2 of the 19 limbs in group 2 patients who were restudied 1 to 2 weeks after having undergone unsuccessful therapy; one patient had complete occlusion of the superficial femoral artery secondary to an attempted balloon

3 630 Kram et al. Journal of VASCULAR SURGERY Table I. Mean Ptco2 values at rest, after 4 minutes of limb ischernia, and at 30-second intervals during recovery Mean Ptco2 (torr) Sensor Subject After arterial position group Control occlusion 0.5 min 1.0 min 1.5 min 2.0 min Chest * 61.9 _ _ a _ b 40.5 _ Calf _ _ a _ _ b Foot _ _ _ a _ b Values were recorded simultaneously at all three levels: chest, calf, and foot. Group 1, normal (N = 22); group 2, PVD (N = 19); group 3a, successful therapy (N = 12); and group 3b, unsuccessful therapy (N = 2). Results are mean -+ 1 SD. *p < 0.05 by paired t test for the difference between control and arterial occlusion in group 1. Table II. Mean limb/chest Ptco2 ratios at rest, after 4 minutes of limb ischemia, and at 30-second intervals during recovery Mean limb~chest Ptco2 ratio (%) Sensor Subject After arterial position group Control occlusion 0.5 rain 1.0 min 1.5 min 2.0 rain Calf Foot _ _ _ * _+ 9.5? ? 50.6 _+ 26.3? 3a _+ 7.1" " ~ b " * * _ * ? ? ~ ? 3a 93.0 _ * 53.9 _+ 28.7t 69.3 _+ 27.6? ? 3b " _ _+ 13.0" " Values were recorded simultaneously at all three levels and the calf/chest and foot/chest Ptco2 ratios calculated. Group 1, normal (N = 22); group 2, peripheral vascular disease (N = 19); group 3a, successful therapy (N = 12); and group 3b, unsuccessful therapy (N = 2). Results are expressed as mean -- 1 SD. Note that the differences between groups increased considerably after temporary limb ischemia. *p < 0.01,?p < 0.001, and +p < for the differences between groups 1 and 2 and groups 3a and 3b by unpaired t test, and between groups 2 and 3a by paired t test. angioplasty and the other had an unsuccessful femoral artery thrombectomy. Both patients in group 3b remained symptomatic after their procedures. Five of the 19 limbs in group 2 did not receive operative therapy because of medical contraindications or inoperable disease. Transcutaneous 02 measurements. The transcutaneous oxygen sensor (Kontron, Roche Inc., Everett, Mass.) used in the study consisted of a Clark-type electrode with a ring anode and a large cathode containing a heating resistor and two calibrated precision thermistors; all components are embedded in epoxy. The sensor is prepared by applying a drop of electrolyte solution and a previously cut membrane supplied by the manufacturer. The membrane is secured in place with a snap-on ring. The sensor characteristics include a 90% re- sponse time of less than 20 seconds and less than 3% nonlinearity (Cutan P02 monitor operating manual; Kontron, Roche Inc.). Transeutaneous oxygen recovery half-time (TORT). Measurement of postocclusive data based solely on limb Ptc02 values were found to be quite variable because of variations in total body hemodynamics and systemic oxygen delivery during and after cuff inflation as reflected by variations in chest Ptc02 (Table I). Therefore the limb/chest Ptc02 ratio, previously shown to reflect limb perfusion independent of variations in total systemic oxygen delivery, 1"2 was used to assess limb perfusion during the measurement of postocclusive TORT. In essence, limb Ptc02 was expressed as a percentage of the simultaneously recorded chest Ptc02. Protocol. The sensors were calibrated to room

4 632 [(ram et al. Journal of VASCULAR SURGERY Colf Foot -- TORT, s! I Group I Group Tl" p<o.o01 p<o.oi p<o.ool [] normal [] peripherol vasculor diseose Group "rrf a [] successful therapy Group 111 b unsuccessful theropy TIME (minutes) q p< o.ool p<o.ol q < 0,o01 _J Fig. 3. TORT in normal volunteer subjects, patients with significant PVD, in patients who had successful therapy, and in patients who had unsuccessful therapy. Values are shown as mean -+ SEM. Note that TORT values of limbs with successful therapy were only slightly greater than those of normal subjects, probably reflecting residual disease. val for values of TORT in normal healthy volunteer subjects at both calf and foot positions was determined to establish the normal range of the test. RESULTS Normal subjects Limb and chest Pte02. Chest Ptc02 values in normal subjects (group 1) increased significantly (p < 0.05), whereas the calf and foot Ptco2 values markedly decreased during temporary limb ischemia (Table I). The chest Ptco2 values remained above preocclusion levels throughout the remainder of the observation period (Fig, 1). There were considerable variations in the control limb Ptco2 values and limb/chest Ptco2 ratios, and frequently these values were indistinguishable from those of patients with symptomatic PVD. Both calf and foot resting Ptco2 values decreased markedly during cuff inflation and then returned to preocclusion levels within 1.5 minutes after cuff deflation. Limb Ptco2 continued to rise and usually remained above preocclusion levels throughout the remainder of the observation period (Table I). Transcutaneous oxygen recovery half-time. Normal subjects had strikingly similar postocclusive TORT values, which were uniformly less than or equal to 1.5 minutes at both calf and foot positions (Fig. 2). The mean calf and foot TORT values were (SEM) and minutes, respectively. On the basis of these data, calf TORT values greater than 1.4 and foot TORT values greater than 1.7 minutes should be considered pathologic, since they are outside the 95% confidence level for normal subjects. Patients with PVD Limb and chest Ptco. Patients with symptomatic PVD (group 2) had appreciably lower chest Ptco2 values than normal subjects throughout the observation period. Resting calf and foot Ptco2 values in symptomatic limbs were usually somewhat lower than those seen in normal subjects (Table I). Both calf and foot resting Ptco2 values decreased during cuffinflation to levels similar to levels seen in normal subjects, but the values of patients with PVD usually returned to preocclusion levels after a duration of 3.0 minutes following cuffdeflation (Table I). The mean control limb/chest Ptcoe ratio was significantly lower in patients with PVD than in normal subjects (p < 0.01), but equivocal values were often seen (Table II). Transcutaneous oxygen recovery half-time. Both calf and foot TORT values in symptomatic limbs were significantly longer in duration than those seen in normal limbs (p < 0.001) (Fig. 3). The mean calf and foot TORT value of limbs in group 2 was and (SEM) minutes, respectively. Patients with PVD who had successful therapy Limb and chest PtcO2. Patients who underwent successful therapy for symptomatic PVD (group 3a) had appreciably lower control chest Ptco2 values postoperatively (Table I). Moreover, the increase in chest Ptco2 during temporary limb ischemia was not as great as that seen preoperatively (Fig, 1). This may be a reflection of limited cardiorespiratory functional compensations in patients with PVD.

5 Volume 1 Number 5 September 1984 Assessment of PVD by postocclusive TORT 631 Mean Ptco~ (tort) 2.5 min 3.0 min 3.5 min 4.0 min 4.5 min 5.0 min 64.3 ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± _~ ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± _ ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± _ ± ± ± 6.6 Mean limb/chejt Ptco2 ratio (%) 2.5 rain 3.0 min 3.5 rain 4.0 min 4.5 min 5.0 min 94.5 _ ± _ ± _+ 28.2t t 72.2 _+ 25.8t 74.3 ± 20.9t 77.3 ± 20.2t t 90.7 _+ 14.6t 94.7 ± 13.7" 96.5 ± 13.8" " 98.7 ± 13.5" 98.2 ± 12.6t 66.8 ± ± ± _ ± ± ± ± ± ± ± ± 24.8t 50.6 ± 25.2t 51.6 ± 26.1t 57.3 ± 25.8t 55.8 ± 24.8t 61.6 ± 24.6t 84.6 ± 25.6t 88.7 _+ 25.9t 90.5 ± 25.7* 90.1 ± 24.2* 92.2 _+ 25.8* 91.6 ± ± 12.5t 20.6 ± 12.4t 23.9 ± 16.1t 29.8 ± 24.4t 32.1 ± 27.4* 34.9 ± 31.4" air and then placed on the anterior chest, medial midcalf, and dorsum of the foot. Areas were shaved if necessary and cleansed with an alcohol pad. Sensors were affixed to the skin with the double-sided adhesive rings and contact gel supplied by the manufacturer. Leads were taped to the leg 2 inches above the sensor. A standard pneumatic blood pressure cuff with an inflatable bladder 8 cm wide and of sufficient length to encircle the limb was loosely applied below the knee. The distance from the lower border of the cuff to the calf sensor was at least 4 inches to ensure adequate sensor contact during cuff inflation. Sensors were allowed to equilibrate with the subject in the supine position for 10 minutes. The pneumatic cuff was then inflated to 50 mm Hg above the systolic arterial pressure for a period of 4 minutes and then rapidly deflated. This duration of ischemia was used because oxygen stores in skin and muscle become exhausted in less than 4 minutes after circu- latory arrest# In no case did cuff inflation prove intolerable to the patient or produce detectable arterial injury; however, two patients complained of leg pain during cuff inflation. Simultaneous chest, calf, and foot Ptco~ values were recorded (1) during the control period, (2) after 4 minutes of cuff inflation, and (3) at 30-second intervals for 5 minutes after cuff deflation. Calibration was rechecked at the end of each study. The Ptco2 response to temporary limb ischemia was determined by the TORT, defined as the time required to recover half of the decrease in the limb/chest Ptcoz ratio caused by temporary limb ischemia. Data analysis. The significance of the differences of postocclusive TORT in normal vs. symptomatic limbs and after successful vs. unsuccessful therapy was determined by unpaired Student's t test. The significance of the improvement ha postocclusive TORT after successful therapy was determined by paired Student's t test. The 95% confidence inter-

6 Volume l Number 5 September 1984 Assessment of PVD by postocclusive TORT 633 Control calf and foot Ptco2 values usually improved postoperatively in limbs that received successful therapy, but this increase was frequently absent and not statistically significant. Both calf and foot resting Ptco~ values decreased during cuff inflation to levels similar to those seen in normal subjects. However, they took longer, that is, 2 to 2.5 minutes, to return to preocclusion levels after cuff deflation (Table I). Patients in group 3a usually had significant improvement in their control limb/chest Ptcoz ratio postoperatively (p < 0.05) (Table II). However, many patients had no improvement or worsening of this variable after surgery, even though symptoms had resolved (Fig. 4). Transeutaneous oxygen recovery half-time. TORT values after successful therapy tended to be slightly greater than those seen in normal subjects (Fig. 2), probably reflecting some degree of residual disease. The mean calf and foot TORT value of limbs in group 3a was and (SEM) minutes, respectively. These values represented significant improvement over their preoperative measurements (p < 0.01) (Fig. 3). In both calf and foot positions, the postocclusive TORT was clearly a more sensitive measurement of limb perfusion than the absolute Ptco2 value or limb/chest Ptco2 ratio. Patients with PVD who had unsuccessful therapy Limb and chest Ptco2. The mean chest Ptco2 values in patients that underwent unsuccessful therapy (group 3b) were appreciably lower throughout the observation period than those of all the other groups (Table I). However, there were only two patients in this group and the differences were not statistically significant. In the limbs of two patients who had unsuccessful therapy, the resting Ptco2 measurements improved only slightly in one patient postoperatively and worsened in the other. Both calf and foot resting Ptco2 values decreased during cuff inflation to levels similar to those of normal subjects, but then they returned to preocclusion levels after 3.5 minutes following cuff deflation (Table I). The mean calf/chest Ptco2 ratio at rest in patients who received unsuccessful therapy was indistinguishable from that seen in patients that received successful therapy (Table 2). Transcutaneous oxygen recovery half-time. The mean calf and foot TORT value of limbs in group 3b was and (SEM) minutes, respectively. These values were significantly % CALF Patient A(normal) I00 A--A--~--A- ^ A ^.'~ Patient B(o~ter bypass surgery} A".,o o-o v5.//;/ J,,/.' z5 kl /./ %0., atient B(before bypass surgery) I00 - FOOT Patient A (normal) ~/ a--~--~--zx-.~. E,A"P=otient B (after-bypass surgery) o ' ' i Control Arteriai Occlusion //7 tient B (before bypass minutes after arterial occlusion Fig. 4. Effect of temporary arterial occlusion in normal subject (PatientA) and in patient with intermittent claudication secondary to complete occlusion of the superficial femoral artery (Patient B). Calf and foot Ptco2 values are expressed as percentage of simultaneously recorded chest Ptco2. Note that in patient B, resting limb perfusion was normal prior to operative therapy but postocclusive TORT was considerably delayed. After successful therapy, resting limb Ptco2 values were decreased despite complete resolution of symptoms. However, postocclusive TORT showed marked improvement after successful therapy. longer in duration than those seen in limbs that received successful therapy (p < 0.001). Calf TORT vs. foot TORT In normal subjects (group 1) foot TORT values uniformly equaled or exceeded corresponding calf TORT values. This was as expected, since the distance from the point of arterial occlusion to the Ptco2 sensor was greater in the former. However, this was not the case in limbs with symptomatic PVD, where calf TORT values occasionally were greater than corresponding foot TORT values. This suggests variations in segmental limb perfusion due to collateral circulation and unevenly distributed disease. DISCUSSION The increased chest Ptco2 after cuff occlusion of the lower extremity may reflect increased arterial Po2 secondary to increased minute ventilation stimulated by cuff inflation or they may be attributable to reflex

7 634 Kram et al. Journal of VASCULAR SURGERY blood flow changes. However, no grossly apparent increase in either respiratory rate or tidal volume with cuff inflation was observed. The lower chest Ptc02 in PVD patients probably reflects decreased systemic oxygen delivery from systemic atherosclerosis and reduced cardlorespiratory function, which is commonly seen in patients with PVD. Although the chest Ptc02 increased during cuff inflation and remained higher than preocclusion levels throughout the remainder of the observation period in patients with PVD (Fig. 1), this increase was not as great (nor as significant) as that seen in normal subjects (Table I). Stressing limb circulation by temporary ischemia exaggerates reduced limb perfusion and oxygenation secondary to stenotic arterial lesions. Transcutaneous oxygen measurements are uniquely capable of quantifying this episode of reduced tissue perfusion and subsequent recovery. The use of temporary arterial occlusion as a method for stressing limb circulation allows differentiation of the perfusion recovery in normal vs. symptomatic limbs (p < 0.001) (Table II). The elevated TORT values seen in limbs with symptomatic PVD result from the diminished oxygen supply relative to oxygen demand during occlusive stress and recovery. In both calf and foot positions the postocclusive TORT was clearly a more sensitive and more specific test than the absolute Ptc02 value or limb/chest Ptc02 ratio. Lower extremity blood flow in a resting subject is usually fairly stable, but marked increases in flow may be produced by temporary limb ischemia. The maximum limb blood flow response probably occurs immediately after release of occlusion, but in the past, technical difficulties have often led to underestimation of early flow measurements. 6 Lower extremity oxygenation, as assessed by Ptco~ monitoring, appears to have a different timecourse after temporary limb ischemia. During cuff occlusion there is minimal tissue oxygenation, indicated by extremely low Ptc02 values, but immediately after release of occlusion, it returns toward normal in a roughly exponential fashion. The time constant of this return in limb oxygenation appears to be relatively large, thereby allowing for accurate, repeated assessment of limb perfusion throughout the examination period. Our data indicate that the time-course of postischcmic oxygenation is considerably delayed in patients with PVD. Thus measurement of postocclusive TORT can be used as a noninvasive method to assess PVD before and after surgical correction and in the diagnosis of patients suspected of having pseudoclaudication. Measurement of postocclusive TORT should be performed simultaneously at both calf and foot locations because the combination of both sensor positions may provide important diagnostic information. For instance, an elevated calf TORT combined with a normal foot TORT suggests the presence of significant proximal arterial occlusive disease with a well-developed collateral circulation. Similarly, significant postoperative improvement of the foot TORT without improvement in calf TORT may be attributed to the topography of the bypassing vessel or prosthesis. A normal calf TORT combined with an elevated foot TORT suggests the presence of significant perfusion deficiency from distal disease. An elevated calf TORT as well as an elevated foot TORT suggest the presence of proximal arterial occlusive disease and impairment of overall limb perfusion, respectively. CONCLUSION A noninvasive measurement, postocclusive TORT, that can be used to objectively quantify tissue perfusion defects in patients with suspected PVD has been described. Repeated measurements of TORT may be used to evaluate the presence, extent, and progression of changes in limb perfusion before and after operative therapy. Since the measurement of postocclusive TORT is easy to perform, noninvasive, and does not require effort by the patient, it is a practical and feasible method for repeatedly assessing the severity and progression of PVD. REFERENCES 1. Hanser CJ, Shoemaker WC. Use of a transcutaneous PO2 regional perfusion index to quantify tissue perfusion in peripheral vascular disease. Ann Surg 1983; 197: Kram HB, Shoemaker WC. Use of transcutaneous O~ monitoring in the intraoperative management of severe peripheral vascular disease. Crit Care Med 1983; 11: Kram HB, Shoemaker WC. Diagnosis of major peripheral arterial trauma by transcutaneous oxygen monitoring. Am J Surg 1984; 147: Hummel BW, Hummel BA, Mowbry A, et al. Reactive hyperemia vs. treadmill exercise testing in arterial disease. Arch Surg 1979; 113: Dornhorst AC. Hyperaemia induced by exercise and ischaemia. Br Med Bull 1963; 19: Barcroft H. The mechanism of vasodilation in the limbs during and after arrest of the circulation. Angiology 1972; 10:595-9.