Optimum compress temperature for wound hemostasis

JOURNAL

OF SURGICAL

Optimum

RESEARCH

%,570-573

Compress

(1979)

Temperature

for Wound Hemostasis

RICHARD

T. NIEMCZURA,

M.D., AND RALPH G. DEPALMA,

M.D.’

Department

of Surgery, Case Western Reserve University School of Medicine Affiliated Hospitals, Cleveland, Ohio 44106

and

Presented at the Annual Meeting of the Association for Academic Surgery, Cleveland, Ohio, November 12- 15, 1978 Compress temperatures to hasten hemostasis in surgically incised wounds have been studied. On the basis of the present data, we recommend the use of warm compresses at a temperature of 49°C to accelerate hemostasis in surgically incised wounds. Cold applications to open surgical wounds should be avoided.

It has long been believed that cold applications to bleeding wounds hasten hemostasis. Hippocrates [4] advocated that cold be applied above bleeding sites, undoubtedly influencing modern medical and lay recommendations for cold wound applications. In bleeding from small vessels, not requiring actual ligature, compression is known to be effective in producing hemostasis. The deliberate use of different moist compress temperatures, either above or below body temperature, is controversial. Surprisingly, this simple point has not been carefully documented in the surgical literature. Few objective data support a firm advocacy of either cold or hot direct wound applications to hasten hemostasis. Contrary to popular belief, recent studies [4-61 using a constant flow of cool water over small stab wounds documented a prolongation of bleeding time. The present study was done to determine an approximate optimum, moist compress temperature which would accelerate hemostasis in surgically incised wounds.

kg were anesthetized with ketamine and xylazine. Their abdomens were shaved and the animals were placed in the supine position; six symmetrically placed incisions were outlined with a surgical marking pen. All incisions were made 3 cm in length, three incisions to each side of the midline, and at least 2 cm apart. To initiate each bleeding trial the skin and subcutaneous tissues were incised with a No. 15 blade to the level of the deep fascia. As the skin retracted an oval wound was produced which measured about 3 cm in length and 1.8-2.0 cm in central diameter. Then 2 x 2 gauze sponges moistened in normal saline at temperatures of 4, 15, 25, 37, or 49°C were immediately placed in each wound with gentle manual compression. The compress temperatures selected for investigation were those easily duplicated in clinical settings: saline kept in an ice bath (4”); temperatures just below or above ambient room settings (15 and 25°C); body temperature at 37°C; and 49°C (120”F), a temperature conveniently attained in the operating room by heating sterile saline in METHOD an autoclave. Since thermal injury to the skin occurs with prolonged exposure to temStandard surgical wounds and an animal peratures of 60°C or higher [ 11, an upper bleeding model were developed. Nine male temperature limit of 49°C was chosen as a New Zealand rabbits weighing 3.1 to 4.4 practical safe limit. The sequences of com1 To whom requests for reprints should be ad- press temperatures were varied randomly dressed: 2065 Adelbert Road, Cleveland, Ohio 44106. from animal to animal. In every instance 0022-4804/79/050570-04$1.00/O Copyright 0 1979 by Academic Press, Inc. All rights of reproduction in any form reserved.

570

NIEMCZURA

AND DEPALMA:

571

WOUND HEMOSTASIS

TABLE 1 AVERAGE

BLEEDING

TIME WITH EACH COMPRESS TEMPERATURE AND INTERGROUP FOR SIGNIFICANCE USING STUDENT’S f TEST

ANALYSIS

Temperature (“C)

Bleeding time (set + SD)

4

15

25

37

49

542 2 51

245 k 59

154 + 44

123 2 32

96 c 25

P < 0.005

Statistical comparisons P < 0.05

4 vs 15 4 vs 25 4 vs 49

NSP > 0.05

25 vs 49

25 vs 37 37 vs 49

the compresses were changed at 20-set in- below body temperature tended to prolong tervals, with care taken not to abrade the wound bleeding. wound. The bleeding times of these treated DISCUSSION wounds were recorded as ending with the last compress exhibiting grossly visible The results of this study suggest that temblood. perature-dependent coagulation reactions RESULTS are enhanced as body temperature is apThe results are summarized in Table 1. proached. This is contrary to the recomIn general, bleeding time decreased with each mended use of cold advocated in standard increment in compress temperature. Figure references [7-91. Hemostasis occurs in three closely inter1 illustrates data in all animals showing the variation between individual animals. The related steps: vascular constriction, platelet averaged data are shown graphically in plug formation, and blood coagulation. Each Fig. 2, with a minimum bleeding time at step occurs in a coordinated sequence difcompress temperature of 49°C. All trials ficult to separate in viva [4]. The first step, vascular constriction, occonfirmed the observation that compresses 600 3404ao2 : : z ” 5 c

420360

-

300240160-

I60

120-

I20

60’

60 4

IS TEMPERATURE

25

37 I’C

49

)

I. Bleeding time in all animals and compress temperature. Note individual variation. FIG.

, 4

I IS TEMPERATURE

1 23

I 37 1-C

I 49

I

FIG. 2. Average bleeding time k SD at different compress temperatures. Note minimal bleeding time at 49°C.

572

JOURNAL

OF SURGICAL

RESEARCH:

curs immediately after injury. Cold clearly produces vasoconstriction. This phenomenon is widely known and probably accounts for the advocacy of cold compresses. Less well known, however, is that after the initial vasoconstrictive response to cold, a reactive hyperemia can develop. As noted previously, Hippocrates [4] recommended application of cold above a bleeding site; this step in some circumstances might cause intense distal vasoconstriction, e.g., in an extremity the resulting decreased blood flow would lead to enhanced wound hemostasis. Endothelial adherence of platelets also reportedly increases with low temperature [7], but the exact effects of direct cold applications have not been elucidated in any detail upon platelet function in viva. Important platelet functions are depressed by lower temperature [2]; platelet morphology changes with decreasing temperature; certain platelet factors are not released below 20°C. Even ADP added to pooled platelet-rich plasma fails to cause aggregation at low temperatures. These data support the view that warmer temperatures are overall probably more conducive to normal platelet function in vivo. It has been known since 1772 [l], that cold prolongs the clotting time of whole blood in vitro. This is consistent with the view of blood coagulation as the result of a cascading series of enzymatic reactions [3] culminating in the formation of fibrin. The optimum temperature of many enzymatic reactions is 37”C, including that of at least one, prothrombin, in the coagulation series. Direct studies in man support the use of warmth to hasten hemostasis. Sutor et al. [4-61 determined the effects of temperature on the standard Ivy bleeding time. They employed a l-mm stab wound on the forearm, subsequently producing a continuous flow of water at varying temperatures over these small wounds. The persistence of blood in the eluted fluid was used to determine the endpoint and the quantity of bleeding. They showed clearly an optimum

VOL. 26, NO. 5, MAY

1979

of 37°C at the wound surface for most rapid hemostasis. Clot lysis occurred at wound temperatures over 435°C. These small wounds however differed from the surgically incised larger wound open to ambient environmental temperatures. The data of Willman and Hamlin [9] involved a large raw wound created with a dermatome and used temperatures of much greater extremes, i.e., 65°C for hot compresses. The one set of experiments with compresses of 30°C showed no difference from their “optimal level” of 5°C. Larger wounds do not possess mechanisms to conserve heat; the surface temperature of open surgical wounds very rapidly approaches that of the environment. While surface wound temperatures were not measured in the present study, it is probable that the surface temperature of the warm compress-treated wounds rapidly fell as soon as the applied compresses were handled and applied. Clearly, warmer compresses hastened hemostasis probably by enhancing coagulation enzymatic reactions. This effect overrides any potential vasoconstrictor effect of cold, at least down to temperatures of 4°C. While there were no statistical differences between compresses at 37 and 49”C, there appeared to be rapid cooling of all moist compresses after application. Slightly accelerated hemostasis appeared in each instance with the sponge at 49°C. This temperature was therefore chosen for compresses in the operating room allowing for some cooling of the compress as it is handled and applied. Further study of actual surface temperatures to accelerate hemostasis in a variety of surgical wounds and circumstances will be of great use in minimizing blood loss. REFERENCES 1. Artz, C. P., Yarbrough, D. E. Bums. In D. C. Sabiston, Jr. (Ed.), Textbook of Surgery, 10th ed. Philadelphia: Saunders, 1972. P. 276. 2. O’Brien, J. R. Effect of cold on bleeding. Luncet 1: 89, 1971.

NIEMCZURA

AND DEPALMA:

3. Ratnoff, 0. D. Some recent advances in the study of hemostasis. Circ. Res. 35: 1, 1974. 4. Sutor, A. H. Effect of cold on bleeding: Hippocrates vindicated. Lancet 2: 1084, 1970. 5. Sutor, A. H., Bowie, E. J. W., and Owen, C A. Effect of temperature on hemostasis: A cold tolerance test. Blur 22: 27, 1970. 6. Sutor, A. H., Bowie, E. J. W., and Owen, C. A. Quantitative bleeding time (Hemorrhagometry). Mayo Clin. Proc. 52: 238, April 1977.

WOUND HEMOSTASIS

573

7. Schwartz, S. I., and Troup, S. B. Hemostasis, surgical bleeding and transfusion. In S. I. Schwartz et al. (Eds.), Principles ofSurgery, 2nd ed. New York: McGraw-Hill, 1974. 8. Vistnes, L. Grafting of skin. Surg. C/in. North Amer. 57: 948, 1977.

9. Willman, V., and Hamlin, C. R. Influence of temperature on surface bleeding: Favorable effects of local hypothermia. Ann. Surg. 143: 660, 1956.