USE of indwelling arterial cannulas has become standard practice for monitoring of systemic arterial pressure and sampling of arterial blood in critically ill patients.1 A recent article implicates clots formed at the tip of a Scribner shunt as being responsible for cerebrovascular accidents when these clots were flushed retrograde into the central arterial circulation.2 That study included a calculation of the theoretical volume of flush solution necessary to cause cerebral embolization, but had no confirmatory measurements. The occurrence of sudden unexplained neurologic deficits in patients with indwelling radial arterial cannulas prompted the present study. The purpose was to determine a . . .

There is great practical need for a dependable method of assessing effects of analgesics on experimental pain in man. Past use of threshold pain has not provided such a method. The submaximum effort tourniquet technique was developed to produce experimental pain in man which simulates the duration and severity of pathologic pain more closely than earlier experimental methods. With this method, ischemic pain is produced by having the subject squeeze a hand spring-exerciser 20 times after a tourniquet is inflated around his upper arm. Performance is measured in terms of elapsed time between cessation of squeezing and report of each of four levels of pain : slight, moderately distressing, very distressing and unbearable pain. Pain increases progressively after cessation of squeezing and is terminated by releasing the tourniquet when the subject reports unbearable pain. In the present study, pain was induced in each of 36 healthy male subjects on each of three occasions : once after administration of 15 mg of morphine, once after 7.5 mg of morphine, and once after placebo. The three medications were administered i.v. and in a double blind, counterbalanced order. At each pain level, pain developed more slowly after administration of 7.5 mg of morphine than after placebo, and developed still more slowly after 15 mg. Analyses of variance showed that differences among means for placebo, 7.5 mg morphine and 15 mg morphine were significant at the .001 level for unbearable, very distressing and moderately distressing pain, and were significant at the .05 level for slight pain. The 7.5-mg dosage of morphine prolonged pain tolerance about 13% beyond that found with placebo, and the 15-mg dosage prolonged it about 36%. Linearity of drug effect was highly significant. These results reaffirm and extend our earlier conclusion that pain produced by the submaximum effort tourniquet technique responds dependably to 10 mg of i.v. morphine.