THE COLLAPSE PRODUCED BY VENOUS CONGESTION OF THE EXTREMITIES OR BY VENESECTION FOLLOWIXG CER- TAIN HYPOTENSIVE AGENTS 1 Bu EDIVARD D. FREIS, JOSEPH R. STAXTON, FRANK A. FINNERTY, JR., HAROLD `I'lr. SCHNAPER, ROBERT L. JOHNSON," CHARLES E. RATH, AXD ROBERT W. WILKINS . (From fllc Cavdiokasrl~lar Research Laboratory, Georgeform C;nivcysity Hospital, md Depart- wmt of Medicine, Georgetown lY~&wsity Sciiool of Jledicine, Washi,tgtot~, D. C.; P'efeyuns Admir~istmtion Hospital, Washington, D. C.; Evans Nemovial, Massa- chsetfs illernoriuf Hospitals, and fJze Dep~yfF~z~~kt of Medicine, Bosion UmbersitS School of Medicirle, Bostorl, Mass.) (Submitted for publication December 14, 1950 ; accepted, February 25, 3951) It `is well recognized that interruption of the louver thoracic and lumbar sympathetic nerves, or the administration of certain hypotensive drugs particularly sodium nitrite and the "sympatholytic" agents mav cause h~pote~lsion and collapse in the erect posiiion. It has been postulated that this hyp+tensive reaction is due to pooling of blood in the dependent parts of the body, particularly in the splanchnic vascular bed with consequent faiIure of venous return and, hence, of cardiac output (1) . During studies on the hemodynamic effects of hypotensive drugs in man we have observed sud- den and marked reductions of arterial pressure frequently with collapse occurring in the supine position. This phenomenon was induced by sub- jecting the limbs to brief periods of venous con- gestion, after the administration of certain hypo- tensive agents (2). This paper presents data on the types of agents that cause this plleilolllenon as well as an analysis of the mechanisms whereby the effect is produced. &fATERIALS AND hfETHODS The subjects were patients both hypertensive arid nor- motensive from the hospital wards. Arterial pressure was recorded either by the usual auscultatory method or optically with a Hamilton manometer connected to an 18 gauge needle inserted into the brachial artery. With the patient in the supine position, inflatable cuffs connected to a large air reservoir bottle were applied proximally to both thighs and one arm and were held in place by covering the cuffs and adjacent skin with gauze bandage. In 911 of the hypertensive patients the limbs 1 Supported in part by research grants from the Na- tional Heart Institute, U. S. Public Health Service, E. R. Squibb and Sons, Kew York, N. Y., and Irrvin, `E;eisIer and Company, Decatur, Ill. ZCaptain, M.C., U.S.A.F. were congested with a pressure of 300 mm. Hg and in the normotensive subjects at pressures of 60 to 80 mm. Hg. &fore and after the administration of each of the various drugs tested the extremities were congested for five min- utes unless collapse supervened. After each drug the blood pressure was recorded frequently. The point at which the blood pressure leveled off or began to rise was taken as the point for applying venous congestion. A similar procedure was used in the venesection experiments. The volume of blood trapped in the congested ex- tremities was determined according to the method of Ebert and Stead (3) following a congestion period of five minutes, instead of seven to 10 minutes as used by those investigators.. The shorter congestion period may account for the fact that in the present experiments the volume of blood trapped in the extremities usually was smaller than that observed by Ebert and Stead. Following the congesting period the pressure in the cuffs was raised to 250 mm. Hg after which the dye T-1824 was injected. Samples were drawn at eight and 10 minutes afterwards. The occluding cuffs were then released and further sampling was carried out at 18 and 20 minutes. The latter samples were used to determine total blood volume while the former were used to deter- mine the volume of that portion of the body not removed from the general circulation by the occluding cuffs. It was found that the dye density values of the eight and 10 minute samples agreed closely with each other indi- cating that mixing was essentially complete at the end of eight minutes. Similarly the 18 minute were in good agree- ment with the 20 minute samples. The results of other investigations (4-7) also indicate that in normal sub- jects equilibration of dye is complete` in less than eight minutes. Hence, it seems probable that this method was su~cie~~tly accurate for the present purposes. The segment plethysmograph used to determine changes in calf volume was the same as that described by Wilkins and Eichna (8) with the exception that proximal and dis- tal blood pressure cuffs were not applied since in the present espe'riments the plethysmograph was used to de- termine volume changes rather than blood flow. In the venesection esperiments a 15 gauge needle was inserted into an antecubital vein and the blood introduced 435 436 FREIS, STAKTON, Fih'NERTY, SCHN.%PER, JOHi'XOW, KATH, AIiD WILKISS into a vacuum Aask containing 130 cc. of acid citrate des- trose solution (.4CD). Immediately following comple- tion of the bleeding period the flask was inverted, and rhen pressurized by raising it and introducing air into it in order to hasten the return of blood to the circulation. gestion of the estrernities. Three of these sub- jects developed marked hypotension and symptoms of coll&se within ix-o to three minutes after the congesting pressures were applied. Five subjects were given 0.5 mg. of dihydro- ergocornine (DHO) intravenously. During con- gestion the changes in average arterial pressure in these patients varied between i 6 and - 48 per cent (mean - 27 per cent). Three of these suh- jects developed signs of collapse prior to the end of the five minute congesting period. In three subjects given tetraeth~lanli17oniullll (250 to 400 mg. intravenously) none developed collapse during congestion of the limbs although all ex- hibited reductions in average arterial pressure varying from - 4 to - 22 per cent (mean - 15 per cent). However, all of the four subjects given the longer acting and more potent ganglionic block- ing agent hexamethonium (C,) developed col- lapse within two to five minutes of venous con- gestion. The reduction of arterial pressure in this group of patients varied between - 16 and - 56 per cent (mean - 37 per cent). RESULTS In the control period, prior to the administra- tion of any drug only two of the 24 subjects es- hibited a significant reduction of arterial pres- sure during congestion of the extremities for a five minute period. In. these txo cases (G. S. and T. L., Table I) the reductions in average systolic 5 diastolic 2 ! arterial pressure were 10 and 12 per cent, respectively. Neither of these patients exhibited signs of collapse. Five patients who received sodium nitrite in doses of 90 to 210 mg. by mouth exhibited changes in average arterial pressure varying from + 1 to - 44 per cent (mean - 25 per cent) during con- TABLE I The eflect oj oariorcs hy~ote?zsi~~ agents ox the arterd presswe during con&on of !he esfrenrities of unsympathectomhed patients Control Post-drug period I- - t ^- - Change mean arterial xessure Pt. Age Drug Dose and route - - - i _- - After live min. of congestion mm. Hg l.iS/lOO 98:66t WiO: 68:40 . 176/106 ,;g 108456 ::`i:; 127~kx9 220/142 100:62 SO/60 90:so G/is 75055 192jllO :::;;i" 162/100 1461100 l&5/131 lY6/114 Basal -- mm. Hz 1801106 17oi115 178/99 96150 liO/106 184194 174190 1881104 130/82 114/56 160:106 236/146 lZS/80 160/100 i28jlOO 1 lOjB0 150/l 10 2ooi1to tZS!lDD 138/100 irr;ix,6 lSS/lli 200/l 12 I Basal mm. HE 200/120 2261128 2251109 118/68 lSO/llO 218/120 194/110 204/104 :;o"i;; 19ijl20 2%/142 140/80 ;$?j:;; : liO/llO liO/ltS 23ljll8 130396 1a0/104 210!128 272/180 2651160 25.5/142 I .- - , I - *CT CellI -11 -44 -43 -26 4-1 -48 24: -2s -23 -19 12: -56 -2.5 -16 -50 1," -8 $i - 1 R. M. A. A. E. C. hf. s. A. P. Ii. M. J. S. J. L. G. S. A. J. T. B. k:$ ;i. t: E. s. B. G. C. B. J.&y. 24 48 43 :; 2: 29 28 43 49 52 28 42 49 i": 54 2i :`i 56 52 52 Sodium nitrite Sodium nitrite Sodium nitrite Sodium nitrite giix%rn nitrite ~~~ Es TEAC TEAC F" C6 C6 CS Sodium amytal Sodium amytal Sodium amytal veratrone Veracrone l'ertavis Vertavis IS0 mg. P.O. 120 mg. P.O. 90 mg. P.O. 180 mg. P.O. 210 mg. P.O. 0.5 mg. I.V. 0.5 mg. I.V. OS mg. t.v. 0.5 mg. I.V. 0.5 mg. I.V. 300 mg. I.V. 250 mg. I.V. 400 mg. I.V. 50 mg. I.V. 50 mg. t.v. 50 mg. I.V. so mg. I.V. 0.25 gm. I.V. 0.75 gm. I.V. 0.5 gm. I.V. 0.8 cc. I.M. 1.0 CC. I.M. 40 CRAW V.PO. 40 CRAW V.PO. mm. Hg lQOjll2 2461146 210/t 18 114164 ::bi::t :;4i:z 111!79 114/78 2121133 2521149 118/7S 160/100 1641126 155/120 160/120 2201l23 lx?/92 176/108 204/124 276\182 266/16S 252/149 * "hlean" arterial pressure = svstolic + diastolic oressure 2 f Ccltapx ooxrred 2: minutes after application of congesting pressure. 1 Collapse occurred two minutes after application of congesting pressure. Subject J. G.* S. K.* M. S.* R. M.* 2,990 4,230 1,240 I. P. "CONGESTION-CGLLAPSE'~ AFTER CERTAIN WYPOTENSIVE AGEXTS 437 TABLE 11 The amount of blood trapped in the extremities subjected to venous congestion before and after sodium nitrite, dihydroergocokne and hexamethonium =I: I 4,150 4,040 5,540 Control 5,480 5,075 6,660 Blood trapped in congested limbs 1,330 1,025 1,120 = I After drug Drug dose and route DHO 0.5 mg. I.V. DHO 0.5 mg. I.V. Sodium nitrite 210 mg. P.O. Sodium nitrite 180 mg. P.O. Hexamethonium 50 mg. I.V. cc. 4,860 4,070 4,055 3,010 5,450 Total blood vofume 6$0 5,460 l,TiO 1,390 5,020 965 4,280 1,270 6,540 1,090 BlOOd trapped in co;rb;d * Data on the changes in arterial pressure given in Table I. No significant reductions of arterial pressure of faintness and nausea. Release of the congesting occurred during congestion of the extremities of pressure at this time and tifting the patient into a three patients given 0.25 to 0.75 gm, of sodium five degree head-down position failed to relieve her amytal intravenously, two subjects given 0.8 or symptoms. The arterial pressure continued to I.0 cc, of Veratrone (tincture of veratrum viride) fall and one minute after releasing the pressure in intramuscularly and two patients given 40 Craw the cuffs it was 48/33 mm. Hg and the heart : units of Vertavis (whole Powdered root and rhi- rate was 58 per minute. Four minutes after re- zomes of veratrum viride) orally. In all of the leasing the congesting pressure the arterial pres- latter cases significant reductions of basal arterial sure had risen to 110/78 mm. Hg and the heart pressure had occurred following the administra- rate to 72 per minute at which time the patient tion of veratrum viride. felt improved. The signs and symptoms related to the collapse state induced by sodium nitrite and the "sympatho- Jytic" agents during congestion of the extremities resembled those observed in other types of acute hypotension such as carotid sinus and v&o-vagal syncope, or postural hypotension. The pulse pres- sure became small and bradycardia frequently de- veloped. Pallor, yawning, sweating and vomiting sometimes occurred. The patients experienced faintness, giddiness and at times nausea. For example, patient E. C. (Table I) who re-, ceived 180 mg. of sodium nitrite by mouth 30 minutes prior to congestion exhibited a marked reduction of arterial pressure from 178/100 to 88,' 70 mm. Hg within three minutes after the venous tourniquets had been applied, Accompanying the hypotensive response there was profuse sweating, Pallor and bradycardia. The patient complained In one hypertensive and four normotensive sub- jects the amount of blood trapped in the congested Iimbs was determined by the dye method of Ebert and Stead (3). In the control period the amount of blood present in both Iower extremities and one upper extremity during congestion varied between 1,025 and 1,650 cc. with a mean of 1,273 cc. (Table II, Figure 1). There were no significant changes in arterial pressure. Two subjects received sodium nitrite, two were given DHO and one received hex- amethonium. During the congestion period fol- lowing these drugs all of the subjects exhibited sig- nificant reductions of arterial pressure and three developed marked hy~tension and symptoms of collapse (Table II). The latter occurred two to 438 FREIS, STANTON, FINNERTY, SCHNAPER, JOHNSON, RATH, AND WILKINS FIG. 1. CHART OF ARTERIAL PRESSURE, INCREASE IN VOLUNE OF CALF .%Cbzq AMOUNT OF BLOOD TRAPPED IN THE EXTREMITIES AND CONGESTING CUFF PRESSURFS~ BEFORE ASD AFTER INTRAVENOUS HEXAMETHONIUN (50 mg.) IN SUBJECT I.P., NORMAL MALE, AGED 28 In the control period the arterial pressure was well maintained during venous congestion of the extremities for 5% minutes. Following hexamethonium and 2% min- utes after the venous tourniquets had been applied, hypo- tension and collapse developed. However, the amount of blood trapped in the congested extremities was less and the increase in volume of the calf segment was not as great as in the control period. four minutes after the beginning of congestion. This necessitated application of occluding pressure prior to the end of the five minute congestion pe- riod. This was done in order to prevent blood from either entering or leaving the limbs while the patient was tilted into a head-down position and allowed to recover during the time that the blood volume determinations were being com- pleted. The amount of blood present in the congested limbs during the experimental period varied be- tween 965 and 1,540 cc. (mean 1,263 cc.). In two subjects there was a slight increase in the amount of blood trapped in the limbs in the post-drug as compared to the pre-drug congestion period and in three cases a slight decrease. In no case did the pre-and post-drug values vary from each other by more than 12 per cent. Changes in calf volzr?ne during venozls congestion of the extremities In order to corroborate these observations indi- cating that no more blood is trapped in the limbs during venous congestion after as compared with before the administration of agents which induce "congestion collapse," the volume changes in the calf were measured using a plethysmograph. Hexamethonium was administered to all of the subjects. In the control period the increase in volume of the calf segment during four and one-half to six and one-half minutes of venous congestion varied between 2.4 to 5.2 cc. per 100 cc. of limb volume (mean 3.8 cc.) (Table III, Figure 1). Hypoten- sion did not occur in any of these subjects. After hexamethonium five of the six subjects developed severe hypotension during periods of congestion of one and one-half to four minutes. The increase in calf volume varied between 2.0 and 3.6 cc. (mean 2.7 cc.). Subject P. G. did not develop severe hypotension during application of venous tourniquets for five and one-half minutes. The rate of increase in calf volume during con- gestion usually was greater after as compared to before hexamethonium (Figure 1). However, in TABLE III The increase in volume of the calf se ment during application of zvenous tourniquets before and after the a mznastralzon of hexamethonium 2.. I I Control I After hexamethonium (50 mg. 1.V.) I Increase in volume of calf segment cc. $?er 100 cc. limb volume 3.0 4.0 ;*i 3:o 2.4 Duration of Deo';tIo;pyt Increase in Duration of "enow volume of venous congestion tension calf segment congestion cc. grr 100 cc. limb volume 2.8 3.0 :.i 3:6 2.0 "COKGESTI~N-COLLAPSE" AFTER CERTAIN XYPOTENSIVE AGENTS 439 TABLE IV The e#ect of dikydruergocornine (DHO), hexamdkonium (CJ, and sadilrm nitrite on the arterial pressure and henrt rate foBowing venesection and subsequent return of the remooed blood to the tircul.ation Pt. L. H. J. B. M. D. M.C. E. A. E. R. w. 0. .P %: and route cs 50 mg. I.V. CS 50 mg. I.V. DHO K mg- * 0 NaN03 Fg mg . . Before drug Arterial pressure mm. Hg 170~100 220/f 20 130/100 2201160 L22f82 240/140 190/M Heart rate B- min. 82 88 84 80 94 60 80 = Afterrcd&u~ and venesection = I Arterial pressure mm. Hg 140/80 140/90 130j100 2201140 122182 180/120 195/112 Heart rate P*r min. 86 84 84 80 96 84 80 Blood re- moved 170 200 300 400 500 100 200 300 400 500 100 200 300 400 500 100 200 2.50 100 200 300 400 100 200 300 400 525 150 300 400 500 During venesection * Yawning, pallor, feels faint. four of the five patients who developed hypoten- was 1,270 cc. Thus, the excess of blood pooled CT22 I Heart rate w min. g"; 100 100 80 84 80 80 :i 80 92 ;: 104 80 Fix 92 ii 104 84 During return of removed blood Biood h&d iii 200 300 400 500 100 200 350 450 500 100 200 300 400 500 12.5 250 120 200 300 400 52.5 150 400 500 135f95 145flOO t:$:;i 160/120 sion during the post-drug congestion period marked reduction in arterial pressure occurred before the t0tn.i increase in calf volume was as great as that observed at the end of the pre-drug congestion period. The estimation of the amount of blood trapped in the congested limbs as determined by the ple- thysmographic method agrees approximately with that obtained by the dye method. Previous studies estimate that about 750 cc. of blood are normally Present in thk lower extremities (3). The average amount of blood found in the congested extremities out was approximately 1,270 - 750 or 520 cc. Us- ing the plethysmographic method the average in- crease in calf volume during the control period was approximately 5.6 cc. per 100 cc. of limb volume. If the volume of both lower extremities is estimated to be approximately 10 liters the amount of blood pooled out would be 5.6 x 100 or 560 cc. It should be emphasized that this is a crude estimation since the plethysmograph measures the volume change in only one segment of the limb which may repre- sent a selective rather than a representative por- tion of the vascular bed being congested. 440 FREIS,. STANTON, FINNERTY, SCHNAPER, JOHNSON, RATH, AND WILKINS Effects of certain hypotensive drugs on the arterial untreated subjects beyond which "compensation" pressure during venesection fails and the arterial pressure falls suddenly. Seven patients, of whom five were hypertensive, As a control the arterial pressures of eight blood were bled rapidly froni an antecubital vein with donors were measured during venesection. No removal of 250 to 525 cc. of blood (Table IV). pretreatment with hypotensive agents had been Two patients received 0.5 mg. of dihydroergocor- given. The removal of 500 cc. of blood at a similar nine intravenously, two were given 50 mg. of hexa- rate from each of the donors resulted in a decrease methonium by the same route and three received of average arterial pressure of - 8 to + 2 per cent 120 to 240 mg. of sodium nitrite orally. All of (average - 3 per cent), these patients exhibited reductions of average ar- terial pressure varying from - 10 to - 42 per cent; in five patients the reduction exceeded - 24 per cent. Two patients developed signs of col- lapse after withdrawal of 250 and 500 cc. of blood, respectively. The arterial pressure did not fall precipitously _ at any point in the bleeding period. Instead there usually was a stepwise decrease with each 100 cc. of blood removed (Table IV and Figure 2). Simi- larly on returning the blood to the circulation the increase in arterial pressure paralleled the rate at which the blood was reinfused. Thus, there was no evidence of a critical amount of blood loss as in Absence of congestion collapse after hypotensive drugs in partially sympathectowized subjects Eight hypertensive patients who had undergone either a lumbodorsal, or a transthoracic splanch- nicectomy four months to five years previously were studied. Two patients received sodium ni- trite in doses of 150 to 180 mg. orally, one rel ceived 0.3 mg. of the mixed dihydrogenated alka- loids of ergot (CCK) intravenously and five were given 50 mg. of C, intravenously (Table V). In only two subjects were significant decreases in average arterial pressure observed (- 11 and - 22 per cent, respectively) during the post drug con- ART6o RIAL PRESSURE-MM HG 1 120. 60. BLOOD REMOVED BY VENESECTION AND RETRANSFUSED-CO, 0 too 200 300 400 500 I 0 5 IO MINUTES 15 20 25 30 - 1 FIG. 2. CHART OF ARTERIAL PRESSURE AND THE AMOUNT OF BLOOD REMOVED BY VENESECTION AND &TRANSFUSED IN PATIENT J. B., FEMALE, AGED 48 Hexamethonium (50 mg.) was given intravenously 30 minutes prior to be- ginning the venesection. Yawning, pallor and a feeling of faintness developed when the blood pressure fell to its lowest point. Pt. M. C. P. D. "CONGESTION-COLLAPSE" AFTER CERTAIN HYPOTENSIVE AGENTS TABLE V The effect of sodium nitrite, Ca and CCK on the arterial +ressure during congestion of the extremities of sympathectomized patients 38 36 ii 35 47 38 - = _- , L 1 I I I I I - Drug Z-Z -- %= route kdium 150 mg. P.O. nitrite Sodium 180 mg. P.O. nitrite ;iK 0.3 mg. 1,V. zs 50Ymg. I.V. v. 50;mg. I.V. 2 50,mg. I.V. 3 SOTmg. I.V. *a 50jmg. I.V. 441 Post-drug period mm. Eg per cm1 164/100 0 212/128 215/135 j-3 :;~~:;~ %~E c: 2201140 205/140 -5 2101138 -1 160/108 -2 146194 225/13.5 ERGO 140/120 170/125 124195 mm. Hg 1601105 134190 ~~~~~~" 130/100 140/120 140/120 90160 -7 0 2; 0 :: 0 0 -11 -22 I I: * "Mean" arterial pressure = systolic -f- diastolic pressure 2 . gestion period. Neither of these subjects de- veloped signs or symptoms of collapse. Six of the eight had significant reductions of basal arterial pressure following the hypotensive agents and all of these exhibited increased postural hypotension. DISCUSSION The deveIopment of collapse during venous con- gestion of the extremities has been observed in normal subjects (3) and in patients who have suffered recent blood loss (9). Our observations indicated, however, that patients given certain hy- potensive agents are unusually susceptible to the development of collapse produced by venous tour- niquets and/or hemorrhage. The type of hypo- tensive agents producing this phenomenon were those which also cause postural hypotension (10, ll), that is, sodium nitrite, hexamethonium, tetra- ethylammonium and the dihydrogenated alkaloids of ergot; whereas hypotensive drugs such as vera- trum viride and sodium amytal which do not ordi- narily cause postural hypotension also did not in- duce congestion collapse. The increased susceptibility. to collapse during venous congestion following sodium nitrite and the sympatholytic agents concei