Life-threatening intrathoracic complications during treatment with extracorporeal membrane oxygenation

Journal of Pediatric Surgery VOL XXIII, NO 7

JULY 1988

Life-Threatening Intrathoracic Complications During Treatment With Extracorporeal Membrane Oxygenation By Joseph B. Zwischenberger, Robert E. Cilley, Ronald B. Hirschl, Kurt F. Heiss, Vincent R. Conti, and Robert H. Bartlett Galveston, Texas and Ann Arbor, Michigan 9 Extracorporeal membrane oxygenation (ECMO) has~ been successful ( > 8 0 % survival) in 35 centers in > 9 0 0 newborns with severe respiratory failure having an estimated mortality of > 8 0 % on conventional management, During the last 3 years w e have treated 79 newborns with 74 survivors (94%). Their diagnoses included meconium aspiration, persistent fetal circulation, respiratory distress syndrome, congenital diaphragmatic hernia, and sepsis. Seven patients (9%) had life-threatening intrathoracic complications requiring emergent intervention while on ECMO: tension hemothorax (3), tension pneumothorax (2), and pericardial tamponade (2). Pericardial tamponade and tension hemothorax and pneumothorax show a similar pathophysiology of increasing intrapericardial pressure and decreasing venous return. Perfusion is initially maintained by the nonpulsatile flow of the ECMO circuit before further decrease in venous return results in decreasing ECMO flow and progressive hemodynamic deterioration. Each of the seven patients demonstrated a clinical triad that includes increasing PaO 2 and decreasing peripheral perfusion (as evidenced by decreasing pulse pressure and decreasing SvO2) followed by decreasing ECMO flow with progressive deterioration, The diagnoses w e r e confirmed by transillumination, chest x-ray, or cardiac echocardiogram. Initial emergent placement of a percutaneous drainage catheter was temporizing in all seven cases. However, four patients required emergent thoracotomy for definitive treatment while still on ECMO. All seven patients were weaned from ECMO and are short-term survivors (6 months to 3.5 years). As use of ECMO for newborn severe respiratory failure increases, responsible physicians must be familiar with life-threatening intrathoracic complications and appropriate treatment strategies. 9 1988 by Grune & Stratton. Inc. INDEX WORDS: Extracorporeal membrane oxygenation (ECMO); tension pneumothorax; tension hemothorax; pericardial tamponade.

X T R A C O R P O R E A L membrane oxygenation (ECMO) is the term used to describe prolonged extracorporeal cardiopulmonary bypass achieved by extrathoracic vascular cannulation. Extracorporeal membrane oxygenation has been successfully used (>80% survival) in 35 centers on over 900 newborns with severe respiratory failure having an estimated

E

Journal of Pediatric Surgery, Vol 23, No 7 (July), 1988: pp 599-604

mortality of >80% on conventional ventilator and pharmacologic management? Despite this apparent success, life-threatening intrathoracic complications can occur during ECMO that contribute to the morbidity and mortality with the technique. We describe the clinical setting and pathophysiology and suggest treatment strategies for these problems. MATERIALS AND METHODS From January 1, 1984, through March 31, 1987, we used E C M O "to treat 79 infants less than 10 days of age with severe respiratory failure. Seventy-four (94%) of the patients survived. The current report retrospectively evaluates seven of these 79 patients (9%) who had life-threatening intrathoracic complications requiring emergent intervention while on ECMO. Three presented with tension hemothorax, two with tension pneumothorax, and two with pericardial tamponade. Their diagnoses included meconium aspiration, persistent fetal circulation, respiratory distress syndrome, congenital diaphragmatic hernia, and sepsis. Standard venoarteria! ECMO as described by Bartlett et al 2"H was used on all patients except one, in whom single cannula venovenous ECMO was initially attempted. When support was inadequate, he too was converted to standard venoarterial ECMO. All life-threatening intrathoracic complications described occurred while patients were on venoarterial ECMO. The technique of ECMO consists of prolonged extracorporeal cardiopulmonary bypass by extrathoracic vascular cannulation of the internal jugular vein and common carotid artery to achieve venoarterial bypass. A modified heart-lung machine is used consisting of a small venous blood drainage reservoir, a servo-regulated roller pump that stops when venous drainage is inadequate, a Sci-Med Kolobow spiral coil membrane oxygenator (Sci-Med Life

From the Department of Surgery, The University of Texas Medical Branch, Galveston, and the Department of Surgery, The University of Michigan, Ann Arbor. Presented at the 36th Annual Meeting of the Surgical Section of the American Academy of Pediatrics, New Orleans, October 31 to November 2, 1987. Address reprint requests to Joseph B. Zwischenberger, MD, Assistant Professor, Cardiothoracic Surgery, University of Texas Medical Branch, Galveston, TX 77550. 9 1988 by Grune & Stratton, lnc. 0022-3468/88/2307-0001 $03.00/0 599

600

ZVVISCHENBERGER ET AL

Systems Inc, Minneapolis) to exchange oxygen and carbon dioxide, and a Sci-Med heat exchanger to maintain normothermia. The patients are maintained on continuous heparin anticoagulation titrated to keep the whole-blood activated clotting time (BaSon) 12 between 240 and 280 seconds (1.5 to 2 times normal) to prevent thrombosis of the circuit. Platelets are administered when platelet counts are <80,000. Extracorporeal flow is established to totally support gas exchange (approximately 120 mL/kg/min). To minimize barotrauma to the lungs while the patients are on ECMO, ventilator settings are reduced to "lung rest" conditions (respiratory rate 10, inspired oxygen concentration [FiO~] _<40%,peak inspiratory pressure [PIP] _<20 cmH20). If a chest tube was inserted prior to ECMO, waterseal suction is maintained at 10 cmHzO. When lung recovery begins, extracorporeal flow is decreased and ECMO is discontinued when gas exchange is adequate on low ventilator settings. Patients are maintained alert and awake while on ECMO. Patients on ECMO have systemic arterial pressure recorded continually from either an infrarenal umbilical artery catheter, a posterior tibial artery catheter, or a femoral artery catheter. Patient arterial gases are obtained hourly and transcutaneous oxygen saturation is monitored continuously via a pulse oximeter (Nellcor Inc, Hayward, CA). Since 1985, we have monitored continuous mixed venous oxygen saturation (SvO2) via an Oximetrix fiberoptic catheter (Abbott Critical Care Systems, Mountain View, CA) inserted into the venous return tubing. In addition, ECG and temperature are monitored constantly. Lighting equipment is available in the neonatal intensive care unit for bedside transillumination studies for pneumothorax. Portable chest roentgenograms are available within five to 15 minutes and cardiac two-dimensional echocardiography and pulsed Doppler studies are available 24 hours a day but may require up to an hour response time. RESULTS

Seven patients (9%) h a d life-threatening intrathoracic complications requiring e m e r g e n t intervention: tension h e m o t h o r a x (3), tension p n e u m o t h o r a x (2), Table

1.

Life-Threatening

Complications

a n d p e r i c a r d i a l t a m p o n a d e (2) ( T a b l e 1). E a c h of the seven patients d e m o n s t r a t e d a clinical t r i a d t h a t included increasing PaO2 and d e c r e a s i n g p e r i p h e r a l perfusion (as evidenced by decreases in pulse pressure a n d mixed venous oxygen s a t u r a t i o n ) followed by a decrease in E C M O flow with progressive hemodyn a m i c deterioration. T h e diagnosis of the i n t r a t h o r a c i c complication was confirmed by t r a n s i l l u m i n a t i o n (1) or chest r o e n t g e n o g r a m (5) for tension h e m o t h o r a x and p n e u m o t h o r a x . C a r d i a c e c h o c a r d i o g r a m was used to confirm the diagnosis of p e r i c a r d i a l t a m p o n a d e in both cases. Initial e m e r g e n t p l a c e m e n t of a p e r c u t a n e o u s catheter was t e m p o r i z i n g in all seven cases. F o r tension h e m o t h o r a x and p n e u m o t h o r a x , a no. 12 F r e n c h a r g y l e chest tube was inserted and 10-cm water-seal suction applied. F o r p e r i c a r d i a l t a m p o n a d e , two-dimensional c a r d i a c e c h o c a r d i o g r a m was used to guide p e r c u t a neous p l a c e m e n t of a 14-gauge a n g i o c a t h e t e r for aspiration of blood f r o m the p e r i c a r d i u m to obtain i m m e d i ate relief of s y m p t o m s . Definitive t r e a t m e n t for each of the three patients with tension h e m o t h o r a x u l t i m a t e l y required thoracotomy, exploration, and control of the bleeding site. In two cases, a tension h e m o t h o r a x followed a left anterior t h o r a c o t o m y for p a t e n t d u c t u s arteriosus ligation p e r f o r m e d while on E C M O . U n f o r t u n a t e l y , a m a j o r operative p r o c e d u r e while fully a n t i c o a g u l a t e d has a high risk of postoperative bleeding. One case of tension h e m o t h o r a x with associated p e r i c a r d i a l t a m p o n a d e was p r e s u m e d secondary to t r a u m a during chest t u b e p l a c e m e n t for a p n e u m o t h o r a x prior to E C M O . Both cases of tension p n e u m o t h o r a x occurred in

Requiring Emergent

Intrathoracic

Diagnosis

Weight ( k g )

Sepsis

3.9

Persistent fetal clrcalation

2.7

Meconium aspiration

4.1

Repiratory distress syn-

3.4

drome Meconium aspiration

3.3

Meconium aspiration

3.9

Congenital diaphragmatic hernia

2.6

Complication

Presentation

TPaO2:85~400 ~Perfusion:SvO2 7 5 4 3 0 ~ECMO:450--*200 Tension hemothorax ~'PaOz:8C ~ f 80 ~Perfusion:~pp ~ECMO:35(~ 100 Tensionhemothorax/ TPa02:90~*364 pericardialtamponade ~Perfusion:SvO2 8 2 ~ 3 5 ~ECMO:500--~IO0 Tensionpneumothorax ~PaOz:80--~350 ~Perfusion:~pp ~ECMO:425~ 150 Tensionpneumothorax TPaO=:80--*250 ~Perfusion:SvO2 80--~40 ~ECMO:3SO-~ 100 Pericardialtamponade TPaOz:75~225 ~Perfusion:~pp ~ECMO:420--~200 Pericardialtamponade TPaOz:80--*185 ~Perfusion:~pp ~ECMO:30~0

Diagnostic Study

Tensionhemothrorax

Intervention

Emergent Treatment

While on ECMO Definitive Treatment

Presumed Cause

CXR

Chest tube

Thoractomy

PDA ligation

CXR

Chest tube

Thoracotomy

PDA figation

CXR

Chest tube/needie aspiration

Thoracotomy/pericardial

Chest tube trauma

Chest tube

Chest tube

TransilluminationChesttube CXR

Chest tube

CXR

Cardiac ECHO

Echoguided Angiocatheter

Cardiac ECHO

Echoguided Angiocatheter

drainage

Pediatric feeding tube inserted over guidewire Pericardial drainage; repair of needle perforation of aorta

Occluded chest tube placed pre-ECMO for pneumothorax Occluded chest tube placed pre-ECMO for pneumothorax Injury to right atrium

Percutaneous aspirations of

pneumopericardiumprior to ECMO

NOTE. Pa02: partial pressure of systemicarterial blood [torr). Perfualon: clinical assessment of adequacy of peripheral perfusion. SvO=:mixed venous oxygen saturation (% saturation). ECMO: ECMO flow (mL/min).

Abbreviations: CXR, chest x-ray; PDA, patent ductus arteriosus; pp, pulse pressure; Echo, echocardiogram.

INTRATHORACIC COMPLICATIONS DURING ECMO

patients who had occlusion of chest tubes placed prior to ECMO and were successfully managed with additional chest tube insertion. It is important to recognize that tension pneumothorax occurred despite "rest" ventilator settings in both of these patients. Pericardial tamponade in one patient was definitively managed by placing a no. 5 French pediatric feeding tube with multiple cut sideholes over a guidewire using the angiocatheter placed for emergent decompression. Tamponade was relieved for the duration of ECMO and the tube was removed 24 hours following ECMO. The second patient with pericardial tamponade developed recurrent tamponade 24 hours after initial angiocatheter decompression. Aspiration of the pericardium using the angiocatheter (which had been left in place) yielded drainage of 20 mL/min, temporarily allowing restoration of ECMO flow. When aspiration of blood was no longer possible and severe tamponade recurred with no pulse and no ECMO flow, an emergency left thoracotomy with pericardial drainage was performed. After relief of tamponade, immediate recovery ensued with restoration of ECMO flow and peripheral perfusion. Careful exploration demonstrated an aortic root puncture wound, which was repaired. In retrospect, the puncture wound in the root of the aorta probably occurred prior to ECMO during percutaneous aspirations for pneumopericardium, with late bleeding due to heparinization. All patients with bleeding as the inciting cause of the intrathoracic complication were managed as described by Cilley et al. ~3Patients were maintained at the lowest possible activated clotting time (200 seconds) to prevent circuit thrombosis. Platelet count was maintained at >100,000, and ECMO was discontinued (and heparin stopped) as soon as mechanical ventilatory support could be achieved even if this meant higher ventilator pressures than desired. All seven patients were weaned from ECMO and are short-term survivors (6 months to 3.5 years).

601

catastrophic hemodynamic deterioration. The factors that deserve immediate evaluation when this occurs include venous catheter placement, adequacy of systemic volume status, and the possibility of extracorporeal circuit failure. However, once these potential problems have been eliminated, one should consider intrathoracic complications within the patient that can cause immediate hemodynamic deterioration on ECMO: pericardial tamponade and tension hemothorax or pneumothorax. Each of our seven patients demonstrated a clinical triad that included increasing PaO2 and decreasing peripheral perfusion (as evidenced by decreases in pulse pressure and mixed venous oxygen saturation) followed by a decrease in ECMO flow with progressive hemodynamic deterioration. Continuous monitoring of the patient for immediate detection of the described clinical signs allowed recognition of the developing clinical triad by the ECMO nurse. This prompted the diagnostic procedures and the therapeutic interventions in all seven cases. Pericardial tamponade and tension hemothorax and pneumothorax show a similar pathophysiology of increasing intrapericardial pressure and decreasing

Art Car

Venous Cannula

DISCUSSION

Extracorporeal membrane oxygenation is a technique that has been available to the clinician for the last 10 years. 1424 It has been used successfully in 35 centers on over 900 newborns with severe respiratory failure having an estimated mortality of >80% on conventional management.23 We have reported significant success with >90% survival in the most recent cases n45 treating meconium aspiration, persistent fetal circulation, infant respiratory distress syndrome, and congenital diaphragmatic hernia. While on venoarterial bypass it is unusual to have

Fig 1. Normal physiology on ECMO. Because of poor native lung gas exchange, there is a physiologic right-to-left shunt. This causes desaturation of the mixed arterial blood, which is the sum of ECMO flow plus native cardiac output.

602

Fig 2. Physiology of increased intrathoracic pressure while on ECMO. W i t h decreased venous return to the heart, pulmonary blood flow is decreased, the native cardiac output is decreased, and the relative contribution of the E C M O circuit (postoxygenator POa > 300 torr) to peripheral perfusion is increased, The peripheral arterial partial pressure of oxygen (PaOa) will increase, while the patient will actually have decreased peripheral perfusion.

venous return. Perfusion initially is maintained by the nonpulsatile flow of the ECMO circuit before further decreases in venous return result in a decrease of ECMO flow and progressive hernodynamic deterioration (Figs 1 and 2). With decreased venous return to the heart, pulmonary blood flow is decreased, the native cardiac output is decreased, and the relative contribution of the extracorporeal circuit to peripheral perfusion is increased. Therefore, peripheral perfusion is initially maintained by the nonpulsatile flow of the ECMO circuit (postoxygenator PO2 > 300 torr). The peripheral arterial partial pressure of oxygen (PaO2) will increase while the patient will actually have decreased peripheral perfusion with a decreased pulse pressure and decreased SvO2. Decreased SvO2 confirms the decreased oxygen delivery achieved by the ECMO flow alone. If the problem is not corrected, further progression of pericardial tamponade or tension hemothorax or pneumothorax will result in progressive obstruction of venous return, a decrease in the ECMO flow, and further hemodynamic deterioration of the patient. The diagnosis of tension hemothorax and pneumothorax may be suggested by transillumination of the

ZWISCHENBERGER ET AL

chest but is best confirmed by chest roentgenogram. Extracorporeal membrane oxygenation does not affect the classic appearance of tension hemothorax or pneumothorax on chest roentgenogram. Likewise, pericardial tamponade may be suggested on chest roentgenogram by enlargement of the cardiac silhouette. Most helpful, however, is a cardiac echocardiogram, which will demonstrate a pericardial effusion and may also localize a hemothorax. For emergency treatment of both tension hemothorax and pneumothorax and pericardial tamponade, we recommend placement of a percutaneous drainage catheter to reverse the developing pathophysiology. This was temporizing in all seven of our cases. For pericardial tamponade, placement of an angiocatheter into the pericardium under ultrasound guidance seems most safe and reliable. Once partial drainage using the angiocatheter has relieved the tamponade, a guidewire may be passed using modified Seldinger technique to place a multiholed drainage tube. We have used successfully a no. 5 French pediatric feeding tube for this purpose. For tension hemothorax and pneumothorax a needle, angiocatheter, and chest tube are all options for emergent decompression. If these measures are unsuccessful or the patient initially responds but later hemodynamically deteriorates, emergency thoracotomy may be necessary for drainage of a hemothorax or a hemopericardium. Any time a site of abnormal bleeding is identified during ECMO, the immediate strategy is to modify the anticoagulation.~~ The platetet count is increased to > 100,000, and the heparin dose is decreased to allow the activated clotting time to approach 1.5 times normal (200 seconds). If there is life-threatening hemorrhage, one may consider one to two hours off systemic heparinization. Likewise, it is important to explore immediately a patient who demonstrates evidence of continued bleeding from needle puncture sites, chest tube sites, or a thoracotomy used to ligate a patent ductus arteriosus. All seven of our patients are short-term survivors (6 months to 3.5 years). Although this series of patients is small, the clinical triad of increased PaO2 and decreased peripheral perfusion (as evidenced by decreased pulse pressure and decreased SvO2) followed by decreased ECMO flow with progressive hemodynamic deterioration has consistently appeared when tension hemothorax, pneumothorax or pericardial tamponade have occurred on ECMO. As use of ECMO for newborn severe respiratory failure increases, responsible physicians must be familiar with life-threatening intrathoracic complications and appropriate treatment strategies.

INTRATHORACIC COMPLICATIONS DURING ECMO

603

REFERENCES

1. ECMO National Registry. University of Michigan, 1987 2. Bartlett RH, Gazzaniga AB: Extracorporeal circulation for cardiopulmonary failure. Cnrr Probl Surg 15:1-96, 1978 3. German JC, Worcester C, Gazzaniga AB, et al: Technical aspects in the management of the meconium aspiration syndrome with extracorporeal circulation. J Pediatr Surg 15:378-383, 1980 4. Klein MD, Andrews AF, Bartlett RH: Extracorporeal membrane oxygenation for newborn respiratory failure, in Clinical Aspects of Perinatal Medicine. New York, Macmillan, 1985, pp 232-248 5. Nugent J: Extracorporeal membrane oxygenation in the neonate. Neonatal Network 4:27-38, April 1986 6. Slata M: Extracorporeal membrane oxygenation support of the infant. Dimens Crit Care Nurs 1:70-79, 1982 7. Toomasian JM, Haiduc N J, Wetmore NE, et al: Refinements in prolonged extracorporeal membrane oxygenation in children and neonates. J Extracorp Technol 11:109-118, 1979 8. Wetmore NE, Bartlett RH, Gazzaniga AB, et al: Extracorporeal membrane oxygenation (ECMO): A team approach in critical care and life support research. Heart Lung 8:288-295, 1979 9. Zwischenberger JB, Bartlett RH: Extracorporeal circulation for respiratory or cardiac failure, in Livetta JM, Taylor RW, Kirby RR (eds): Critical Care. Philadelphia, Lippincott (in press) 10. Zwischenberger JB, Cilley RE, Andrews AF, et al: The role of extracorporeal membrane oxygenation in the management of respiratory failure in the newborn. Respir Care 31:491-497, 1986 11. Toomasian JM, Haiduc N J, Zwischenberger JB, et al: Techniques of extracorporeal membrane oxygenation for cardiac failure. Proc Am Acad Cardiovasc Perfusion 7:105-109, 1986 12. Braden JP, Sonnenfield M, Ferlic RM, et al: The BaSon test: A rapid bedside test for control of heparin therapy. Surg Forum 22:172-173, 1971 13. Cilley RE, Zwischenberger JB, Andrews AF, et al: Intracranial hemorrhage during extracorporeal membrane oxygenation in neonates. Pediatrics 78:699-704, 1986

14. Bartlett RH, Andrews AF, Toomasian JM, et al: Extracorporeal membrane oxygenation for newborn respiratory failure: 45 cases. Surgery 92:425-433, 1982 15. Bartlett RH, Gazzaniga AB, Toomasian JM, et al: Extracorporeal membrane oxygenation (ECMO) in neonatal respiratory failure: 100 cases. Ann Surg 204:236-245, 1986 16. Bartlett RH, Gazzaniga AB, Wetmore NE, et al: Extracorporeal membrane oxygenation (ECMO) in the treatment of cardiac and respiratory failure in children. Trans Am Soc Artif Intern Organs 26:578-579, 1980 17. Bartlett RH, Gazzaniga AB, Huxtable RG, et al: Extracorporeal circulation (ECMO) in neonatal respiratory failure. J Thorac Cardiovasc Surg 74:826-833, 1977 18. Bartlett RH, Roloff DW, Cornell RG, et al: Extracorporeal circulation in neonatal respiratory failure: A prospective randomized study. Pediatrics 76:479-487, 1985 19. Hardesty RL, Griffith BP, Debski RF, et al: Extracorporeal membrane oxygenation: Successful treatment of persistent fetal circulation following repair of congenital diaphragmatic hernia. J Thorac Cardiovasc Surg 81:556-563, 1981 20. Kirkpatrick BV, Krummel TM, Mueller DG, et al: Use of extracorporeal membrane oxygenation for respiratory failure in term infants. Pediatrics 72:872-876, 1983 21. Krummel TM, Greenfield L J, Kirkpatrick BV, et al: Clinical use of an extracorporeal membrane oxygenator in neonatal pulmonary failure. J Pediatr Surg 17:525-531, 1982 22. Towne BH, Lott IT, Hicks DA, et al: Long-term follow-up of infants and children treated with extracorporeal membrane oxygenation (ECMO): A preliminary report. J Ped Surg 20:410-414, 1985 23. Trento A, Griffith BP, Hardesty RL: Extracorporeal membrane oxygenation experience at the University of Pittsburgh. Ann Thorac Surg 42:56-59, 1986 24. Weber TR, Pennington DG, Connors R, et al: Extracorporeal membrane oxygenation for newborn respiratory failure. Ann Thorac Surg 42:529-535, 1986

Discussion Clyde Redmond (New Orleans): 1 think Dr Zwischenberger has very nicely elucidated the pathophysiology of this problem, but in reality, it is just an extension of the normal physiology of cardiopulmonary bypass. You can see that if you have a patient who has nice PaO2s as measured at the umbilical artery catheter it may be because you have dried the patient out by drawing blood out of the circuit. We have had trouble convincing our neonatologists that what is more important than PaO 2 is venous saturation. Even though the patient may only have a PaO2 of 60 with a venous saturation of 60% or 70%, they are getting adequate oxygen delivery. Mitchell Ross (Denver): Of the seven patients that had these complications, how many of them were postoperative patients, and of those who were not, do you have any explanation as to why they would have

developed hemorrhagic complications in the thorax and mediastinum? Could it possibly be due to the large venous catheter sitting in the superior vena cava or right atrium? Joseph Zwischenberger (closing): I want to thank Dr Redmond for his discussion. As you know, Dr Arensman was one of the pioneers in the use of ECMO, and his experience is well recognized in the field. 1 agree that what I presented here today is an illustration of normal physiology while on E C M O and how increased intrathoracic pressure with a decrease in venous return can produce a cascade of events that causes catastrophic deterioration. I think that Dr Arensman's group in New Orleans and other groups throughout the country should be recognized for their safe conduct of ECMO. Regarding the causes of the intrathoracic bleeding,

604

all of these patients are managed with total anticoagulation with heparin. It is very important that people who do ECMO realize that any surgical wound, any needle puncture site, or any site of traumatic injury is a potential source of bleeding. As is illustrated in the full article, two of the patients who developed hemorrhage

ZWlSCHENBERGER ET AL

probably did so as a result of well-meaning attempts by neonatologists to aspirate pneumothorax or fluid prior to the institution of ECMO. Two tension hemothoraces were the result of patent ductus arteriosus ligation while on ECMO.