Diagnosis and Treatment of Shock Due to Massive Pulmonary Embolism

clinical investigations in critical care Diagnosis and Treatment of Shock Due to Massive Pulmonary Embolism* Approach With Transesophageal Echocardiography and Intrapulmonary Thrombolysis Bojan Krivec, MD; Gorazd Voga, MD, MSc; Ivan Zuran, MD, MSc; Rafael Skale, MD; Roman Pareznik, MD; Matej Podbregar, MD; and Marko Noc, MD, PhD

Study objectives: To evaluate the diagnostic value of transesophageal echocardiography (TEE) as an initial diagnostic tool in shocked patients. The second objective was to study therapeutic impact of intrapulmonary thrombolysis in patients with diagnosed massive pulmonary embolism. Design: Prospective observational study. Setting: Medical ICU in 800-bed general hospital. Patients: Twenty-four consecutive patients with unexplained shock and distended jugular veins. Measurements and main results: In 18 patients, right ventricular dilatation with global or segmental hypokinesis was documented. In addition, central pulmonary thromboemboli (12 patients), reduced contrast flow in right pulmonary artery (one patient), and right ventricular free wall akinesis (one patient) were found. No additional echocardiographic findings were apparent in four patients. According to pulmonary scintigraphy or autopsy, sensitivity of TEE for diagnosis of massive pulmonary embolism (MPE) in patients with right ventricular dilatation was 92% and specificity was 100%. In patients without right ventricular dilatation, left ventricular dysfunction (four patients) or cardiac tamponade (two patients) was confirmed. Intrapulmonary thrombolysis was evaluated in 11 of 13 patients with MPE. Two patients died prior to attempted thrombolysis. Three patients received streptokinase and eight received urokinase. Twenty-four hours after beginning of treatment, total pulmonary resistance index significantly decreased for 59% and mean pulmonary artery pressure for 31%. Cardiac index increased for 74%. Nine of 11 patients receiving thrombolysis survived to hospital discharge. Conclusion: Bedside TEE is a valuable tool for diagnosis of MPE. It enables immediate intrapulmonary thrombolysis, which seems to be an effective therapeutic alternative in our group of patients with obstructive shock. (CHEST 1997; 112:1310-16) Key words: intrapulmonary thrombolysis; massive pulmonary embolism; right heart catheterization; shock; streptokinase; transesophageal echocardiography; urokinase Abbreviations: MPE=massive pulmonary embolism; RPA = right pulmonary artery; TEE=transesophageal echocardiography

Adequate treatment of patients with unexplained shock requires accurate and prompt evaluation of the etiology and assessment of cardiac function. Unfortunately, appropriate diagnostic procedures *From pepartment of Intensive Internal Medicine (Drs. Krivec, Voga, Zuran, Skale, Pareznik, and Podbregar), General Hospital Celje, Celje, Slovenia, and the Center of Intensive Internal Medicine (Dr. Noc ), University Clinical Centre, Ljubljana, Slovenia. Manuscript received July 17, 1996; revision accepted May 2, 1997. Reprint requests: Bojan Krivec, MD, Dept of Intensive Internal Medicine, General Hospital Ceije, Oblakova 5, Ceije 3000, Slat;enia 1310

For editorial comment see page 1158

may be constrained with logistic problems, including transportation of patients. Therefore, an accurate and bedside noninvasive diagnostic tool that allows concurrent therapeutic interventions in the ICU would be of great importance. There is rapidly growing experience with transesophageal echocardiography (TEE), whose diagnostic accuracy in the ICU has already been documented.1 Despite encouraging reports by Nixdorff Clinical Investigations in Critical Care

et aF and Wittlich et al, 3 •4 this method has not gained wide acceptance for diagnosis of massive pulmonary embolism (MPE). Pulmonary scintigraphy and angiography still remain the preferable diagnostic tools. These methods, however, require transportation of the patient from the ICU, which may, in turn, further aggravate patient condition. When MPE is diagnosed, rapid reversal of pulmonary obstruction becomes the primary objective. Thrombolytic therapy is generally believed to be an effective therapeutic alternative despite the absence of randomized trials that would demonstrate improved survival.S However, the selection and administration of thrombolytic agents remains controversial. Since the consensus points toward shorter and more concentrated regimens, central and rapid administration of thrombolytic therapy was suggested. 6 It was hypothesized that a higher concentration of ~;ombolytic agent at the site of thromboembolus may account for more rapid resolution. Indeed, encouraging results using either right atrial 7 or intrapulmonaryil-9 thrombolysis were reported in small and uncontrolled studies. However, shorter and more concentrated thrombolytic regimens were also more effective when administered systemically. lO Our prospective trial investigated diagnostic accuracy of TEE for diagnosis of MPE in patients with unexplained shock and distended jugular veins. At the same time, we also investigated the effectiveness of intrapulmonary thrombolysis with streptokinase and urokinase.

thermodilution catheter to further delineate the structures and to estimate the flow. Each TEE was recorded by video recorder. If right ventricular dilatation was documented, pulmonary scintigraphy was performed within 24 h. Right ventricular dilatation was diagnosed when diastolic diameter of the right ventricle exceeded diastolic diameter of the left ventricle in trans gastric short -axis view at the transpapillary level. Perfusion defects >30% were considered diagnostic for pulmonaty embolism if no radiographic changes were apparent in the regions with perfusion defects. Autopsy was performed routinely in patients who died. Only patients with MPE were included in further study. The 7F -four-lumen thermodilution catheter (Arrow International; Reading, Pa) was inserted through the percutaneous sheath introducer system in the right subclavian vein. The TEE probe was kept in the esophagus to monitor the advancement of the catheter over the tricuspid valve and its positioning close to the thromboembolus. Hemodynamic measurements included central venous pressure, pulmonary attery pressure, and cardiac output. Cardiac output was obtained by thermodilution method using a monitor (Sirecust Syste m 960/961; Siemens; Iselin, NJ). Three consecutive measurements obtained within 2 min were averaged. Cardiac index was calculated as cardiac output divided by body surface area (Umirtlm 2 ) . Total pulmonary resistance index was calculated as mean pulmonary arterial pressure divided by cardiac index (mm HgXminXm 2 XL - 1 ). Arterial lactate level was measured using colorimetric enzyme method (Chronolab; Hitachi; San Jose, Calif).

Shock with distended jugular veins (n= 24)

Left ventricular dysfunction

Cardiac tamponade

(n=4)

MATERIALS AND METHODS Consecutive patients with unexplained circulatory shock admitted to our ICU from July 1993 to February 1996 were included. The inclusion criteria were clinical signs of peripheral hypoperfusion (anuria/oliguria, peripheral cyanosis, cold and pale skin), systolic BP below 90 mm Hg, and distended jugular veins. None of the patients had previous chest trauma. TEE using 5-MHz monoplane probe (Sonos 1000; HewlettPackard; Andover, Mass) was performed within 15 min after admission and served as initial diagnostic method. Right and left ventricles were examined from transverse gastric short-axis view. The visualization of atria and interatrial septum was obtained by slight withdrawal of the probe. Pulmonaty trunk, proximal segment of left pulmonary artery, and whole right pulmonaty artety (RPA) were examined from upper transverse view using extreme clocbvise rotation of the probe. In the pulmonary artery or its branches, any abnormal intraluminal structure with different echoicity compared to blood and vessel wall was considered thromboembolus. Identification of discrete shape (curvilinear in longitudinal or circular in cross section) and definite border was essential to exclude possible artifacts. Masses were dichotomously qualified as mobile or immobile and homogeneous or heterogeneous according to gray scale brightness of structures (heterogeneous having clear central areas and homogeneous with no such areas). In case of suboptimal visualization, lO mL of agitated lipid emulsion (Lipofundin; Braun; Melsungen, Germany) was injected intrapulmonarily through the distal port of a

(n= 2)

- Dilatative cardiomyopathy (2)

-Aortic dissection (I)

-lschentic cardiomyopathy (2)

-Myocardial rupture in AMI (I)

Right ventricular dilatation (n= 18)

Global hypokinesis

Segmental akinesis

(n= 17)

(n= I)

- Massive pulmonacy embolism

- RV infarction

- central thromboemboli (12 ) - reduced RP A !low ( l ) - Septic shock ( I )

- ARDS due to burns ( I ) - COPD with chronic cor pulmonale ( 2 )

FIGURE l. Echocardiographic findings and final diagnosis in 24 patients with shock and distended jugular veins. AMI=acute myocardial infarction; RV=right ventricle.

CHEST I 112 I 5 I NOVEMBER, 1997

1311

Table !-Characteristics, Treatment, and Outcome of 13 Patients With Massive Pulmonary Embolism* Patient/Sex/ Age, yr

LACT, mM

l/F/85 2/M/57 3/M/52 4/F/84 5/M/78 6/M/74 7/M/49 8/F/74 9/F/76 10/M/71 11/ M/67 12/F/67 13/lvl/59

6.5 5.9 23.0 3.7 5.6 10.0 4.3 11.6 4.3 3.8 4.0 4.4

Symptoms Dys Sy11,dys Syn,dys Dys Dys Dys Syn,dys Dys Dys Dys Dys Dys Dys

Underlying Condition

Fibrinolysis

Heart failure Endarterectomy Hip replacement Immobility, diahetes Anticoagulant withdrawal Osteosynthesis None Immobility Osteosynthesis Immobility Osteosynth esis Diabetic loot Spinal lam inectomy

None ;\lone Strept Strept Strept Urok Urok Urok Urok Urok Urok Urok Urok

Supp01tive Therapy CPR CPR V, D ,~

D,N, MV V,D,l'i,MV V,D,N V,D,N V,D,~

D,N,MV D,N D,N, MV D,N, MV D,\1

Outcome Died Died Discharged Died Died Discharged Discharged Discharged Discharged Discharged Discharged Discharged Discharged

*LACT=initial arterial lactate concentration, dys=dyspn ea; syn =syncope; strept = streptokinase; urok = urokinase; CPR =cardiopulmonary resuscitation; V=volume infusions; D=dobutarnine; N=norepineph1ine; MY = mechanical ventilation.

The first three patie nts received continuous infusion of 100,000 IU streptokinase (Boehringer Laboratories; Ridgefield , Conn) pe r h for 24 h through the distal port of a the rmodilution catheter. In the following eight patients, urokinase (Ukidan Se rono, Rome, Italy) was administered in doses of 1 million IU in 15 min, continued by 1 million IU within another 2 h. Thrombolytic therapy was started within 15 min after the diagnosis of MPE was established by TEE. Following thrombolytic therapy, patie nts received continuous heparin infusion in doses of 800 IU/h with appropriate adjustment to maintain thrombin time between 1.5 to 2.5 of normal value. Norepinephrine and dobutamine were administered within the flrst 2 h in doses of 0.2 to 2.0 11-g/min/kg and 3.5 to 10.0 !Lg/minlkg, respectively. The doses were individually titrated in attempt to achieve mean aJterial pressure >80 mm I-Ig, cardiac index >2 ..5 Umin/m 2 , am] urine output >50

mVh. Repeated h emodynamic measure ments were performed during the first 24 h following the start of the treatment. Initial and 24-h measurements of central venous pressure, mean pulmonary arte1y pressure, total pulmonary resistance index, cardiac index,

and systolic arterial pressure were compared using paired twotail ed Stude nt's t test. The p value of <0.05 was considered as signiflcan t.

RESULTS

The study emolled 24 patients (Fig 1). Right ventricular dilatation with global or segmental hypokinesis and leftward bulging of interatrial septum was documented in 18 patients. In these patients, central thromboemboli (12 patients), reduced contrast flow (one patient), and right ventricular free wall akinesis (one patient) were seen. No additional echocardiographic findings were present in four patients. In patients without right ventricular dilatation, either left ventricular dysfunction (four patients) or cardiac tmnponade (two patients) was confirmed. In 18 patients with right ventricular dilatation,

Table 2-Echocardiographic Findings in Patients With Massive Pulmonary Embolism and Confirmation of Diagnosis* Patient 2

3 4

5 6

7 8 9 10 ll lZ

13

Location of Embolus RPA, LPA PA, RPA+LPA 1 RPA RPA RPA+LPA RPA RPA RPA RPA+LPA PA, RPA RPA RPA None

Twe of' Embolus Immobile, homogeneous Mobile, heterogeneous Immobile, homogeneous Immobile, homogeneous Immobile, homogeneous Immobile, homogeneous Mobile, heterogeneous Immobile, homogeneous Mobile, heterogeneous Immobile, homogeneous+ mobile, heterogeneous Mobile, heterogeneous Immobile, homogeneous

Additional Findings Spontaneous contrast in RA, RV Embolus in oval foramen Spontaneous contrast in RPA

Spontaneous contrast in RPA, thrombus in RA

Low-contrast flow in RP A

Confirmation Autopsy Autopsy Scintigraphy Autopsy Autopsy Scintigraphy Scintigraphy Scintigraphy Scintigraphy Scintigraphy Scintigraphy Scintigraphy Scintigraphy

*LPA= left pulmonaty artery; RA=right atrium; RV=right ventricle. 1 Thromboembolus was continuous and extended from RPA to LPA. 1312

Clinical Investigations in Critical Care

MPE was confirmed by either pulmonmy scintigraphy or autopsy on 13 occasions. There were eight male and five female patients, 49 to 85 years old (Table 1). Symptoms of MPE included dyspnea, syncope, or both. Alierial lactate concentrations on admission ranged from 3.7 to 23.0 mmol!L (mean, 7.3±5.6 mmol!L). According to TEE, pulmonary thromboemboli were located in the RPA, left pulmonary artmy, or main pulmonary artery (Table 2). Two different subgroups of thromboemboli were documented: mobile and heterogeneous with central clear area (Fig 2) and immobile and homogeneous with increased density (Fig 3). Additional findings included low contrast flow (Fig 4), spontaneous contrast in right side of heali or pulmona1y aliery, and thrombus in oval foramen (Fig 5). In one patient with MPE, no central masses but only low-contrast flow in the RP A was seen. When conside1ing the detection of central thromboemboli as a sign of MPE in patients with right ventricular dilatation and shock, sensitivity of TEE was 92% (12 of 13 patients) and specificity was 100% (five of five patients). No additional ventilation scans or angiography were performed in patients without thrornboemboli but in three of them, autopsy revealed no pulmonary thromboemboli. Treatment and outcome of 13 patients with MPE are shown in Table 1. Two patients died owing to refractory bradycardia and asystole before initiation of thrombolytic therapy. Eleven patients received either streptokinase (3 patients) or urokinase (8 patients). Since one of the thromboemboli was always located in the RPA (Table 2), the tip of thermodilution catheter was placed in the RP A. The placement was guided by TEE. Suppoliive treatment included mechanical ventilation, volume infu-

FIGURE 2. Two mobile hypoechoic thromboemboli in distal RPA with clear central area (double arrow). SVC=sup rior vena cava; AO = aorta.

FIGURE 3. Large, immobile, and hyperechoic thromboembolus in the proximal part of RPA (single arrow). Them10dilution cath eter is seen in th e supeJior vena cava (right arrows) and in pulmonmy artery (PA, left arrows).

sions, norepinephrine, and dobutamine. Following this treatment, mean pulmonary aliery pressure decreased from 42 to 29 mm Hg and total pulmonary resistance index decreased from 23 to 9 mm HgXminXm 2 XL - J within 24 h (Table 3). Concurrently, cardiac index increased from 1.9 to 3.3 Umin/m 2 and systolic BP from 80 to 129 mm Hg. However, central venous pressure remained essentially unchanged. Total pulmona1y resistance index rapidly decreased in nine patients who survived. In four of them, the values decreased 45 to 60% within

FIGURE 4. Cessation of contrast flow in the middle segment of RPA (right arrows). Contrast is also seen in the supe1ior vena cava.

CHEST I 112 I 5 I NOVEMBER, 1997

1313

DISCUSSION

5 . Thromboe mbolus obstructing the oval fora men . RA = right atrium; IAS = interatrial septum ; LA = left atrium; TV = tricuspid valve; RV= right ventlicle; LV= left ventricle; sin gle arrow = thermodiluti on cath ete r. FIGU RE

the first hour, whereas in two patients who died, total pulmonary resistance index signiflcantly increased (Fig 6). Of nine patie nts who survived to hospital discharge, seven r eceived p eroral anticoagulation and two received inferior caval fllter. No he morrhage due to thrombolytic th erapy was observed. Only one patient who unde1went dural patch after laminectomy for cauda equina syndrome developed transient postoperative liquorrhea.

Our study demonstrated that TEE is a v ery useful bedside diagnostic tool in the setting of unexplained shock with distended jugular veins. Detection of central thromboemboli, aortic dissection, segmental akinesis, and pericardia! tamponade accounted for specific findings that were essential to establish the correct diagnosis . In addition, TEE provided important information regarding right and left ventricular function in each patient. TEE proved to be particularly useful in patients with right ventricular dilatation and suspected MPE. We were able to visualize pulmona1y trunk and whole right pulmona1y artery along with 1ight pulmonary hilus. Left pulmonary artery was visible only in proximal part and in the hilus . It is important to emphasize that complete RPA visualization was performed only by extreme clockwise rotation of the probe. Injection of contrast further improved visualization of the thrombus and yielded additional information regarding the flow characteristics. TEE also demonstrated that mobile thrombi are heterogeneous and immobile thrombi are homogenous . This is in accordance with the study of Wittlich et aP that described similar features of thromboemboli. Their study also enrolled p atients who did not present in shock. Practical importance of this diversity of thromboe mboli, however, remains unclear. TEE also helped to place the tip of Swan-Ganz catheter to the RPA close to tl1e tl1romboembolus despite low flow due to obstruction. The performance of TEE in

Table 3-Hemodynamic Data *

Patient 2 3 4 5 6 7 8 9 10 11 12 13 Mean SD p

MPAP 0, mm Hg

MPAP 24, mm Hg

57 No 51 33 40 29 38 29 40 47 35 48 57 42 10

D 33 D D 26 31 25 28 34 26 33 24 29 4

D

0.003

TPHl 0, mn• H g ~ rmn xm-

XL -

1

TPRl 24, mm Hg x min X m2 XL- 1

51 No 23 18 17 18 27 15 30 32 18 20 20 23 6

D D 7 D

D 6 8 10 13 13 9 13 6 9 3 < 0.001

Cl 0, U min/ m2

CI 24, U mirt!m2

CVP 0, mm H g

CVP 24, mmHg

1.1 No 2.2 1.8 2.3 1.6 1.4 1.9 1.3 1.5 1.8 2.4 2.8 1.9 0.5

D D 4.5 D D 4.0 3 .8 2.8 2.2 2.6 3.0 2.6 4.1 3.3 0.8

18 No 11 20 13 11 14 9 23 16 15 20 24 16 5

D D 8 D D 13 13 10 18 11 14 12 6 12 4

< 0.001

0.07

SAP 0, mrn Hg

SAP 24, rnm I-Ig

Im Im 90 Irn 60 90 Im 90 60 90 95 50 1m 80 15 < 0.001

D D 150 D D 110 115 125 135 110 135 160 125 129 17

*Includi ng mean pulmonmy arte•y press ure (MPAP), total puln 10muy resistance index (TPRI ), cardiac index (CI), and systolic arteri al pressure (SAP) be fo re initiation (0) and24 h (24) afte r th ro mbolytic t herapy. Systolic arte rial pressu re was measured by noninvasi ve cuff method. l m= immeasurable ; D =patient di ed; No = measureme nt not obtained ; p = probabili ty according to Student t test.

131 4

Clinical Investigations in Critical Care

HOURS 0

5

20

15

10

25

160 +--------r------~r-------~-------+------~

140

~

I

-•-

..--

0:::

a..

1--

·- • · ·PATIENT 4

t

::: J. ~ . .. ·-· · · · · · · · · /- -d- - - ~·.t 'A····

80

··~-.

.

··• ·-PATIENT 3

--<>--PATIENT 5 --PATIENTS -a-PATIENT7 -u--PATIENT 8 ---e--- PATIENT

9

-PATIENT10 -PATIENT11

60

---.::-PATIENT 12

40

-::---PATIENT 13

20 FIGURE 6. Relative changes in total pulmonary resistance index (TPRI) (expressed as percent of admission value) in 11 patients with massive pulmonary embolism who received intrapulmona1y thrombolysis with streptokinase (patien ts 3 to 5; dotted lines) and urokinase (patients 6 to 13; continuous lines).

our group of very sick patients did not adversely affect hemodynamic and respirat01y function. Moreover, TEE rather facilitated our treatment strategy, because no transportation out of the ICU for nuclear and angiographic diagnostic procedures was necessary. Since time is so essential in the treatment of shocked patients, it even seems reasonable to bypass transthoracic echocardiography, which may cause unnecessary delay in establishment of a definite diagnosis. In patients with MPE, we observed significant and rapid reduction of pulmonary resistance and pulmonaiy artery pressure, and significant increases in cardiac output and systolic BP concurrent with thrombolysis. In particular, decreases in total pulmonary resistance index within 1 h were impressive and separated survivors from nonsurvivors. In view of the fact that up to 80% of patients with MPE die within 2 h, 11 rapid thrombolysis has the potential to improve survival significantly. In contrast to total pulmonary resistance index, central venous pressure was essentially unchanged. It is therefore likely that this parameter is not very useful for monitoring of hemodynamic status in patients with MPE. The hemodynamic improvement in our patients is comparable to that observed when urokinase in similar doses was administered either in right atrium 7 or pulmonary

artery. 8 However, it is as yet not clear whether central thrombolysis is more efficient than standard regimens with peripheral administration. In conclusion, TEE is a suitable and accurate diagnostic tool in patients with unexplained shock and distended jugular veins. Because TEE is performed at bedside and is safe, we believe it should be available 24 h in the emergency department and ICU. Our combined approach with TEE and hemodynamic monitoring permits the morphologic diagnosis and the assessment of functional impairment. Consequently, etiologic and supportive treatment can be started and titrated concurrently. Thus, TEE together with intrapulmonary thrombolysis may represent a reasonable diagnostic and therapeutic approach.

REFERENCES 1 Vignon P, Mentec H , Terre S, eta!. Diagnostic accuracy and

therapeutic impact of transthoracic and transesophageal echocardiography in mechanically ventilated patients in the ICU . Chest 1994; 106:1829-34 2 Nixdorff U, Erbel R, Drexler M, e t a!. Detection of thromboe mbolus of the right pulmonary artery by transesophageal two-dimensional echocardiography. Am J Cardiol 1988; 61: 488-89 3 Wittlich N, Erbel R , Eichler A, et al. Detection of central pulmonary artery thromboemboli by transesophageal echoCHEST I 112 I 5 I NOVEMBER, 1997

1315

4 5 6 7

cardiography in patients with severe pulmonary embolism. J Am Soc Echocardiogr 1992; 5:515-24 Erbel R, Drozdz J, Ge J, eta!. Bildgebene verfahren in der kardiologie. Internist 1994; 35:1039-55 Gulba DC, Schmidt C, Borst HG, eta!. Medical compared with surgical treatment for massive pulmonary embolism. Lancet 1994; 343:576-77 Dickie KJ, de Groot WJ, Cooley RN, et a!. Hemodynamic effects of bolus infusion of urokinase in pulmonary thromboembolism. Am Rev Respir Dis 1974; 109:48-56 Pettitpretz P, Simmoneau G, Cerrina J, et a!. Effects of a single bolus of urokinase in patients with life-threatening pulmonary emboli: a descriptive tJial. Circulation 1984; 70: 861-66

1316

8 Gonzalez-Juanatey JR, Valdes L, Amaro A, eta!. Treatment of massive pulmonary embolism with low intrapulmonary dosages of urokinase: short-term angiographic and hemodynamic evolution. Chest 1992; 102:341-46 9 Mohindra SK, Udeani GO. Treatment of massive pulmonary embolism with centrally administered tissue-type plasminogen activator. Ann Emerg Med 1993; 22:1349-52 10 Goldhaber SZ, Kessler CM, Heit JA, et a!. Recombinant tissue-type plasminogen activator versus a novel dosing regimen of urokinase in acute pulmonary embolism: a randomized controlled multicenter trial. J Am Coli Cardiol 1992; 20:24-30 ll Turnier E, Hill JD, Kerth WJ. Massive pulmonary embolism. Am J Surg 1973; 125:611

Clinical Investigations in Critical Care