SIMULTANEOUS CHEST COMPRESSION AND VENTILATION AT HIGH AIRWAY PRESSURE DURING CARDIOPULMONARY RESUSCITATION

175

the rise in P-TG, a marker of platelet aggregation,’ was also significantly reduced. The raised p-TG values before hxmoperfusion may be related to platelet abnormalities in FHF.3 Disseminated intravascular coagulation is known in this condition," and platelet activation by the foreign surface of the shunt and central venous line is also possible. The higher p-TG levels in the PGI2treated group may reflect the higher pre-perfusion serum-creatinine levels that these six patients had (see

table).14 In the hsemoperfusions done without PGI2 platelet losses were less severe than previously,’ perhaps because of improvements in the column. Nevertheless, the severe hypotensive episode with a rise in screen filtration pressure in one patient and the significant rise in p-TG in all six patients suggest that platelet damage is still a problem. No significant heparin-sparing effect due to PGI2 was seen, perhaps because platelet losses were smaller than in the dog experiments. 5,15 Now that the system has been shown to be safe, the effect of serial haemoperfusion with PGI2 infusion on conscious level and survival rate in FHF can be evaluated. This study is part of a research programme into the development of liver support systems supported by the Medical Research Council, and we also thank Smith & Nephew Research Ltd and Wellcome Research Laboratories for their support.

Requests for reprints should be King’s College Hospital, London SE5

addressed 8RX.

to

R.W., Liver Unit,

REFERENCES

BG, Weston MJ, Murray-Lyon IM, Record CO, Williams R. Experience at King’s College Hospital with charcoal hæmoperfusion: Overall results in 37 patients. In: Williams R, Murray-Lyon IM, eds. Artificial

1. Gazzard

liver support. London: Pitman Medical, 1975: 234-41. 2. Silk DBA, Williams R. Experiences in the treatment of fulminant hepatic failure by conservative therapy, charcoal hæmoperfusion and polyacrylonitrile haemodialysis. Int J Artif Organs 1978; 1: 29-33. 3. Weston MJ, Langley PG, Rubin MH, Hanid MA, Mellon PJ, Williams R. Platelet function in fulminant hepatic failure and effect of charcoal hæmoperfusion. Gut 1977; 18: 897-902. 4. Langley PG, Hughes RD, Ton HY, Silk DBA, Williams R. The use of PGI2 in the prevention of adverse platelet reactions during charcoal haemoperfusion. Int J Artif Organs 1979; 2: 207-10. 5. Bunting S, Moncada S, Vane JR, Woods HF, Weston MJ. Prostacyclin improves hemocompatibility during charcoal hemoperfusion. In: Vane JR, Bergstrom S, eds. Prostacyclin. New York: Raven Press, 1979: 146-67. 6. Williams R. The management of liver failure. In: Weatherall DJW, ed. Advanced medicine vol 14. London: Pitman Medical, 1978: 19-29. 7. Swank RL. The screen filtration pressure method in platelet research: significant and interpretation. Ser Hœmatol 1968; 1: 146-67. 8. Ludlam CA, Moore S, Bolton AE, Pepper DS, Cash JD. The release of a human platelet-specific protein measured by radioimmunoassay. Thromb Res 1975; 6: 543-48. 9. Gryglewski RJ, Szeczeklik A, Nizankowski R. Antiplatelet action of intravenous PGI2 in man. Thromb Res 1978; 13: 153-63. 10. Szczeklik A, Gryglewski RJ, Nizankowski R, Musial J, Pieton R, Mruk J. Circulatory and anti-platelet effects of intravenous PGI2 in healthy men. Pharmacol Res Commun 1978; 10: 545-56. 11. Szczeklik A, Nizankowski R, Skawynski S, Szczeklik J, Gluszko P, Gryglewski RJ. Successful therapy of advanced atherosclerosis obliterans with PGI2 Lancet 1979; i: 111-14. 12. Tateson JE, Moncada S, Vane JR. Effects of prostacyclin (PGX) on cyclic AMP concentrations in human platelets. Prostaglandins 1977; 13: 389-97. 13. Rake MD, Parnell G, Flute PT, Williams R. Intravascular coagulation in acute hepatic necrosis. Lancet 1970; i: 533-37. 14. Depperman D, Andrassy K, Seelig H, Ritz E, Post D. Evidence for dependence of &bgr;-thromboglobulin levels on renal function. VII International Congress on Thrombosis and Hæmostasis, 1979: abstr p 416. 15. Woods HF, Ash G, Weston MJ, Bunting S, Moncada S, Vane JR. PGI2 can replace heparin in hæmodialysis in dogs. Lancet 1978; ii: 1075-77.

SIMULTANEOUS CHEST COMPRESSION AND VENTILATION AT HIGH AIRWAY PRESSURE DURING CARDIOPULMONARY RESUSCITATION NISHA CHANDRA MICHAEL RUDIKOFF MYRON L. WEISFELDT Peter Belfer Laboratory for Myocardial Research, Cardiology Division and Department of Medicine, Johns Hopkins Medical Institutions, Baltimore, Maryland 21205, U.S.A.

Summary

blood flow during resuscitation (CPR) results from a rise in intrathoracic pressure rather than from direct heart compression. Intrathoracic pressure was increased by the use of positive-pressure ventilation synchronous with sternal compression in eleven arrested patients who were intubated. A computer system allowed 15-30 s periods of alternation between conventional CPR and "new CPR" (rate of 40/min, 60% compression duration, and simultaneous ventilation at airway pressures from 60 to 110 cm H2O). Compression force was identical with the two methods. New CPR increased mean systolic radial artery pressure significantly from 40·6±4·4 to 53·1±3·9 mm Hg for 14 runs in nine patients. In 15 runs in ten patients an index of carotid flow increased with new CPR to 252% (range 113-643%) of control values. Lowering airway pressure during new CPR lowered flow index and arterial pressure, confirming that these increases with new CPR were due to higher intrathoracic pressure. Thus, increased airway pressure synchronous with sternal compression increases arterial pressure and likely blood flow during CPR in man. However, further studies of potential complications and long-term effects of new CPR, particularly on adequacy of ventilation, are needed before clinical implementation is undertaken. In

most patients cardiopulmonary

INTRODUCTION

et al.1 showed that circulation of blood arrested patient can be maintained by sternal displacement and proposed that blood flow resulted from direct compression of the heart between sternum and spine. This explanation, though widely accepted, has been questioned.2 Our studies in dogs3 indicated that a rise of intrathoracic pressure during sternal displacement leads to circulation of blood, and the demonstration that coughing, without chest compression, can generate sufficient cerebral blood-flow to maintain consciousness in arrested patients4 supports the animal work. In the dog experiments intrathoracic pressure rose during compression, producing nearly identical pressure pulses in all intrathoracic cardiac and vascular structures. Intrathoracic pressure during compression is transmitted almost unchanged into the arteries outside the chest, and right atrial and superior vena cava pressure are also nearly identical to intrathoracic pressure during compression. However, the thin-walled jugular veins probably collapse at the thoracic outlet. The transmission of intrathoracic pressure to the extrathoracic arteries but not the veins establishes the peripheral arterial-venous pressure gradient needed for flow to the carotid vascular bed. Manoeuvres which increase in-

IN

in

an

1960, Jude

176

trathoracic pressure increase carotid flow and pressure. Other investigators had noted the similarity of arterial or aortic pressure and central or left atrial pressure 2,5,6 but did not identify how peripheral blood flow resulted nor the benefit of higher intrathoracic pressure. Since these studies suggested that high intrathoracic pressure during chest compression might increase flow and blood pressure during cardiopulmonary resuscitation (CPR) we explored the administration of high airway pressure simultaneously with chest compression. In the dog, this resulted in a significant increase in carotid flow and pressure for over 5 min and no impairment of gas exchange. Wilder et, a1. had demonstrated improved blood flow in dogs with ventilation synchronous with compression at lower airway pressures, but the mechanism of this benefit was unclear and some of the measurements have been questioned.s,9 Following review of our databy the Johns Hopkins Medical Institutions Committee on Human Investigation, we administered simultaneous compression and ventilation at high airway pressures ("new CPR") for brief periods to arrested patients and compared new CPR with conventional CPR. METHODS

CPR

was

started in the standard way, with manual external

compression, endotracheal intubation, positive-pressure respiration, defibrillation, and radial-artery cannulation and drug therapy as indicated. Six patients had myocardial infarctions, failure, and/or shock, one had a head injury, two had had cardiac arrests outside hospital, and two needed resuscitation for other reasons. After initial resuscitative measures had failed, external cardiac compression was continued with a computerised pneumatic piston chest compression device10 capable of producing programmable patterns of external cardiac compression and ventilation at different airway pressures. For conventional CPR chest compression was at 60/min with 60% of each cycle in compression; ventilation occurred during diastole, lasting 0-45 after every fifth compression; airway pressure was 40-60 cm H2O. These settings were held

over time during new CPR and expressed as a percentage of the control value during conventional CPR for the same time period. This gives the carotid flow index, a measure of relative blood flow, assuming constancy of vessel diameter and probe angle. In each patient at the termination of resuscitation the Doppler instrument baseline zero was identical to the zero obtained with the transducer in position over the artery and the piston off (no flow). This baseline was also identical to the measurement obtained during external chest compression with the probe positioned over an adjacent area free of large vessels. The protocol included the provision that conventional CPR would be immediately resumed if radial artery pressure or flow index fell. The protocol was also interrupted for 5 min to allow for haemodynamic equilibration after administration of any vasoactive agents. No data were included where flow index during conventional CPR before and after new CPR varied more than 10%. Group differences were compared using the paired or unpaired Student’s t test as appropriate. ± indicates SEM.

grated

RESULTS

We studied 16 periods of conventional CPR followed new CPR and return to conventional CPR. We were able to monitor both carotid flow velocity and radial artery pressure in eight patients, carotid flow only in two, and radial artery pressure only in one. No more than two runs are reported in any single patient. During new CPR, airway pressures of 60-110 cm H2O simultaneous with chest compression at 40/min were used. New CPR increased mean systolic radial artery pres-

by

from

significantly (p<0.001) mm Hg in a total of 14 runs in nine patients When only one run (the first) per patient was (fig. 1A). analysed this significance persisted (41-1±6-2 to New CPR increased 33.9±5.1 mm Hg; (p<0.001).

sure

40-6+4-4

to

53-1±3-9

arterial pressure from 33-3±4-2 to 45’9±3-5 mm for 14 runs in nine patients and from 33-7±5-99 to for single runs. Carotid flow 48+4.7 mm Hg (p<0.001) index during new CPR increased in all 15 runs in ten patients. The average flow index during new CPR was 252% of conventional CPR (range 113-643%) (fig. 1B).

mean

Hg

constant, except when new CPR was used. For new CPR chest compression was at 40/min with 60% of each cycle in compression; ventilation was synchronous with the onset of compression and continued throughout compression ; airway pressure was 60-110 cm H2O during compression to atmospheric between compressions, thus allowing intrathoracic pressure to fall and venous return to occur. For 15-30 s periods we changed from conventional CPR to new CPR and back to conventional CPR. In a subgroup of six patients during continuous new CPR for 45-60 s, airway pressure was rapidly (5-10 s) varied from 40 to 110 cm H2O and then returned again to 40 cm. Each pressure setting was maintained for 10-15 s. The peak piston force was adjusted initially to produce approximately 6 cm (2-5 in) of sternal displacement during conventional CPR and then held constant for each patient for both conventional and new CPR. The pneumatic arm of the device was fitted with a potentiometer which recorded piston deflection and permitted calculation of chest wall displace-

falling

ment.

Sternal compression force and displacement were monitored constantly. Mean systolic radial arterial pressure was taken as the mean pressure from the onset of chest compression to the onset of compression release. Bidirectional transcutaneous carotid artery flow velocity was measured with a Parks model 806 bidirectional ultra-sonic Doppler flow meter (8-8mHz). A pencil probe was stabilised at a fixed angle relative to the vessel in the neck. The uncalibrated velocity recording was inte-

Fig. I-Systolic arterial

pressure and carotid flow ventional and new CPR.

during

con-

systolic artery pressure in nine patients (two runs shown patients). Patients 1 and 3, who had the highest pressures generated with conventional CPR, had little increase with new CPR. (B) Carotid arterial flow index during new CPR expressed as a percentage of flow during conventional CPR. Patient numbers are as in fig. l a. (A)

Mean

for five

,

I

177

(e.g., patients 1 and 3 in fig. 1) radial artery presgenerated by conventional CPR alone tended to be the patient was thin chested with a large heart, and high suggesting that sternal displacement during conventional CPR may have resulted in direct compression of the heart. Sternal displacement on new and conventional CPR, compression force being constant, were much the same (4-70 ±0-19 cm on new CPR and 5.00±0-15 cm on conventional CPR). Thus, in the small number of patients for whom sternal displacement seems vital, new CPR was certainly not disadvantageous. However, in the much larger group of patients with lower arterial pressures during conventional CPR increasing airway pressure and thus intrathoracic pressure at constant sternal compression force seemed to be the sole mechanism for the major improvement attained small sures

of flow velocity, radial artery presin one patient showing onset of new CPR and return to conventional CPR.

Fig.

2--Continuous

sure, and

tracing

compression force

CPR. For blood to flow there

on new

venous

Fig. 2 illustrates carotid flow index and radial artery blood-pressure during conventional and new CPR in one patient. In the entire group, patients with the smaller increases in carotid flow index with new CPR were those in whom the higher arterial pressures were generated with conventional CPR (see Discussion). At 60% compression duration, compression-rates of 40 or 60/min are associated with similar values for mean systolic pressure or flow index,1O so the major difference between new and conventional CPR was the high airway pressure synchronous with compression. This was confirmed by lowering airway pressure during new CPR (fig. 3). Flow index at 60-110 cm H20 was 249±63% of the value at 40 cm H2O (range 117-597%, n=6).

DISCUSSION

Our findings indicate that simultaneous compression and ventilation at high airway pressures (new CPR) greatly increased carotid flow and radial artery pressure over values achieved with conventional CPR. Most patients showed a major increase in arterial pressure and flow index with new CPR, and where the changes were

from the to a lower venous pressure outside the chest. Where intrathoracic pressure is transmitted directly to the extrathoracic arteries the necessary gradient is generated. That this mechanism operates during CPR is supported by measurements during Swan-Ganz catheter pullback3 in three patients (where pressure dropped by 12-31 mm Hg as the catheter passed from the intrathoracic veins into the extrathoracic non-obstructed internal jugular vein) and by the higher flow index and arterial pressure brought about by higher airway pressure (fig. 3) and new CPR. Systolic radial artery pressure cannot be equated with intrathoracic pressures during compression since there is likely to be significant damping and/or

amplification. The Doppler technique assumes constancy of vessel diameter and probe angle. However, no runs were included which were done during or after administration of any vasoactive medications, and by including only those runs in which flow index varied no more than 10% before and after new CPR we hoped to limit the data to runs with constancy of probe position. Carotid artery collapse, which we had noted at high airway pressure during CPR in dogs3 was of immediate concern during our studies in man. However, we saw no decreases in pressure or flow index and the increase in flow index was accompanied by increased pressure. Many of the patients had atherosclerosis and were at an age where increased arterial stiffness might resist collapse and none of them, before arrest, had had hypovolasmia, a factor which in the dog seemed to result in

frequent carotid collapse.3 Despite the dramatic increases in flow index and blood-pressure seen with new CPR, we advise against rapid clinical implementation of this approach. The dog studies showed no impairment of oxygenation, but we have no data in man on long-term effects on oxygenation or CO2 removal. Nor have we evidence of long-term benefit in terms of augmentation of flow, and the possibility of carotid collapse during long-term new CPR has yet to be ruled out. Real benefit, in terms of survival or lessened cerebral dysfunction, has not been demonstrated. Although airway pressures are high during new CPR, overexpansion of the lung is prevented by simultaneous compression. However, the technique looks promising, more

Fig. 3-Percutaneous carotid Doppler flow velocity and pulmonary arterial pressure during new CPR (slow paper speed) during lowering and raising of airway pressure from 110 to 40 cm H,O. Brief period of conventional CPR is shown on right. Pressure and forward flow velocity (above zero) rise and fall in relation to changes in airway pressure. Although piston starts from higher point during new CPR, net displacement is similar.

must be a peripheral arterialgradient. This gradient probably results collapse of veins at the thoracic outlet leading

pressure

178

and clinical trials of this unconventional method of CPR seem warranted. We thank the Emergency Care Research Institute (Waltham, Massachusetts) for the card-programmable cardiopulmonary resuscitation equipment, made to their specifications by Michigan Instruments, Inc. (Ann Arbor).

Requests for reprints should be addressed to M. L. W., Johns kins Medical Institutions, Baltimore, Maryland 21205, U.S.A.

Hop-

Supported by grant no. P 50 HL 17655-03 from the National Heart, Lung and Blood Institute, Bethesda, Maryland. N. C. is a fellow of the Maryland Heart Association. REFERENCES

1. Kouwenhoven WB,

Jude JR, Knickerbocker sage. JAMA 1960; 173: 1064-67.

GC. Closed heart cardiac

mas-

Hypothesis IS AUTOIMMUNITY A COMMON DENOMINATOR IN IMMUNE COMPLEX DISEASES? EDMUND

J. LEWIS

JIMMY L. ROBERTS

Section

of Nephrology, Department of Medicine, Rush-Presbyterian-St. Luke’s Medical Center, Chicago, Illinois 60612, U.S.A.

normal circumstances, antibodies reactive with native DNA appear in the plasma during the course of many clinical conditions associated with inflammation. These antibodies seem to be elaborated in response to the release of exceptional amounts of DNA by nucleated cells. As a result, DNA/ anti-DNA complexes can be demonstrated in the cryoprecipitable fraction of plasma from patients with various inflammatory diseases. A significant proportion of these immune complexes contain low-molecular-weight polynucleotide antigens. These polynucleotides are derived from DNA which has been degraded by plasma DNAase. Because of the digestion of DNA in the plasma a spectrum of antigen/antibody complexes forms. While large, relatively insoluble complexes would be expected to be rapidly cleared by the reticuloendothelial system, low-molecular-weight complexes are removed more slowly. It is proposed that the action of plasma DNAase upon both free and immune bound-DNA can lead to a preponderance of small, soluble, polynucleotide/antiDNA complexes. Under appropriate conditions of vascular permeability, these soluble complexes may be deposited in vessel walls. Hence, regardless of the initiating infectious or inflammatory agent, polynucleotide antigen/anti-DNA antibody complexes form and can result in immune-mediated inflammatory phenomena in diverse disease states.

Summary

In

BACKGROUND

ALTHOUGH reactivity of the immunological system with autoantigens is recognised in many forms of tissue damage, the presence of antibodies directed against native DNA antigens is generally believed to be abnor-

2. Weale FE, Rothwell-Jackson RL. The efficiency of cardiac massage. Lancet 1962; i: 990-92. 3. Rudikoff MT, Maughan WL, Effron M, Freund P, Weisfeldt ML. Mechanisms of blood flow during cardiopulmonary resuscitation. Circulation 1977; 56: suppl III, 97 abstr; and Circulation (in press). 4. Criley JM, Blaufuss AN, Kissel GL. Cough-induced cardiac compression.

JAMA 1976; 236: 1246-50. 5. MacKenzie GJ, Taylor SH, McDonald AH, Donald KW. Hæmodynamic effects of external cardiac compression. Lancet 1964; i: 1342-45. 6. Thomson JE, Stenlund RR, Rowe GG. Intracardiac pressures during closed chest cardiac massage. JAMA 1968; 205: 46-48. 7. Chandra N, Rudikoff M, Tsitlik J, Weisfeldt ML. Augmentation of carotid flow during cardiopulmonary resuscitation (CPR) in the dog by simultaneous compression and ventilation with high airway pressure. Am J Cardiol 1979; 43: 422 abstr. 8. Wilder RJ, Weir D, Rush BF, Ravitch MM. Method of coordinating ventilation and closed chest cardiac massage in the dog. Surgery 1963; 53: 186-94. 9. Harris LC, Kirimli B, Safar P. Ventilation-cardiac compression rates and ratios in cardiopulmonary resuscitation. Anesthesiology 1967; 28: 806-12. 10. Taylor GJ, Tucker WM, Greene HL, Rudikoff MT, Weisfeldt ML. Importance of prolonged compression duration during cardiopulmonary resuscitation in man. N Engl J Med 1977; 296: 1515-17.

mal and characteristic of autoimmune disease. The demonstration of B lymphocytes which possess surface aritigen-binding sites for native DNA in normal, subjects has provided evidence that immune reactivity with extracellular DNA may be a normal function.Anti-nativeDNA antibodies and double stranded DNA have been found in the cryoprecipitable fraction of plasma immunoglobulins isolated from patients with diverse inflammatory disorders, including otherwise normal indiwith intercurrent infectious diseases. These observations support the notion that anti-DNA antibody synthesis and the formation of DNA/anti-DNA complexes can occur during the inflammatory process. Production of antibodies to native DNA has been observed in species other than man. In mice antibodies to native and single-stranded DNA appeared in the circulation subsequent to experimental conditions which cause DNA release.3 The concept of normal auto-reactivity to DNA and other nuclear antigens casts a new light upon the importance of the formation of anti-DNA antibodies in clinical situations other than systemic lupus erythematosus (SLE). One is led to reject the concept of a "forbidden clone" of lymphoid cells as the explanation for the appearance of these autoantibodies. Rather, we propose that the elaboration of anti-DNA antibodies may be part of a homoeostatic reaction to the release of DNA into the circulation.

HYPOTHESIS

the formation of immune complexes microbial containing foreign antigens. However, the role of these antigens has not been consistently established in immune deposits. We propose that a common process can underlie the pathogenetic events leading to many antigen/antibody-complex-mediated diseases. Cellular injury from infectious or inflammatory processes resulting would be expected to cause the release of DNA into the circulation. DNA is enzymatically degraded to polynucleotides by plasma DNAase. Small polynucleotides derived from released DNA may not be cleared from the circulation as rapidly as are larger molecules. These polynucleotides stimulate anti-DNA antibody production. These latter immunoglobulins aid removal of polynucleotides from the blood. However, the low-molecInfection leads

to