HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Harris, D. G. Articles by Rossouw, G. J. Search for Related Content PubMed PubMed Citation Articles by Harris, D. G. Articles by Rossouw, G. J.

Ann Thorac Surg 1999;68:2119-2122
© 1999 The Society of Thoracic Surgeons

---

Original Articles: Cardiovascular

Current evaluation of cardiac stab wounds

^a Department of Cardiothoracic Surgery, Tygerberg Hospital, University of Stellenbosch, Cape Town, South Africa

Address reprint requests to Dr Harris, Department of Cardiothoracic Surgery, Tygerberg Hospital, Suite A2, Cape Town, 7505, South Africa
e-mail: dharris{at}maties.sun.ac.za

	Abstract

Top Abstract Introduction Material and methods Results Comment References

 
Background. Patients with penetrating cardiac injuries may be stable or only mildly shocked, especially if the laceration has sealed off and the patient has been aggressively resuscitated. Clinical signs, chest roentgenograms, pericardiocentesis, and subxiphoid window are not always helpful in establishing the diagnosis. We reflect on the current evaluation based on 128 patients.

Methods. There were four groups of patients, ranging from lifeless (group I) to stable (group IV). Patients in groups I and II were prepared immediately for operation. Those in groups III and IV were often investigated further (chest roentgenogram and cardiac ultrasound).

Results. Mortality was 8%. Significant findings were a precordial stab, central venous pressure of more than 15 cm of water, one or more clinical signs of tamponade, and initial shock. Cardiac ultrasound was performed in 5 patients in group II (15%), 14 patients in group III (48%), and 37 patients in group IV (86%). There were no false positives, and 6 false negatives (11%). Thirty-one patients (24%) had clotted lacerations. There were no negative sternotomies.

Conclusions. Efficient fluid resuscitation and rapid confirmation of diagnosis with cardiac ultrasound should decrease mortality. Stable patients with a precordial wound should undergo cardiac ultrasound or echocardiogram. Diagnosis may be reliably confirmed in these patients whose clinical signs often fluctuate (or rapidly deteriorate).

	Introduction

Top Abstract Introduction Material and methods Results Comment References

 
Most surgeons agree that patients with penetrating cardiac wounds need immediate operation. However, some patients appear to be stable, or remain truly stable. Clinical signs may fluctuate. Hypotension may be temporarily corrected by aggressive fluid resuscitation. The central venous pressure (CVP) may be lower than expected because of unrecognized bleeding into the pleural space, or falsely elevated due to pain, straining, shivering, pneumohemothorax, or catheter malposition (or truly elevated by aggressive fluid resuscitation), leading to a false diagnosis of cardiac tamponade. Stable patients may indeed have a cardiac laceration that may remain undiagnosed—until the patient suddenly deteriorates [1]. Investigations, such as chest roentgenogram, pericardiocentesis, and subxiphoid window, are not always reliable. Echocardiography has been useful in stable patients with minimal clinical signs [2]. Cardiac ultrasound has not been extensively used.

Our recent experience is discussed, with particular reference to the more stable patient.

	Material and methods

Top Abstract Introduction Material and methods Results Comment References

 
Between July 1994 and August 1997, 128 patients were referred to our unit with the diagnosis of a penetrating cardiac injury. There were 122 male and 6 female patients. There were 124 stabs and 4 gunshot wounds. The mean age was 29 years (5 to 55 years). Two patients had previously undergone operation for a stabbed heart. Thirty-five patients had previously been treated in our trauma unit for violence-related injuries.

Four groups were recognized after resuscitation: group I, no vital signs (23 patients); group II, agonal (systolic blood pressure [SBP] remains < 90 mm Hg, 33 patients); group III, compensated shock (SBP remains > 90 mm Hg, 29 patients); and group IV, stable (43 patients).

The following protocol has been adopted by our trauma unit: (1) suspicion with all precordial wounds, especially if associated with hypotension and distended neck veins; (2) a decision to proceed further in lifeless patients if some signs of life were noted by the transferring paramedic; (3) immediate insertion of subclavian CVP line and high flow infusion sets on each arm or leg, or saphenous vein cut down. Immediate infusion of 2 L of plasmalyte B solution as well as colloids or O negative blood. Acidosis is corrected; (4) if there are no vital signs, immediate endotracheal intubation is performed and a chest tube is inserted on the side of the injury; (5) if the patient remains lifeless, emergency room thoracotomy is performed; (6) a patient who develops respiratory arrest but still has signs of some circulation is intubated and transferred to the operating room; (7) patients with SBP less than 90 mm Hg with hemorrhage or tamponade are transferred directly to the operating room; and (8) if the patient has a SBP more than 90 mm Hg after resuscitation, a mobile chest roentgenogram and a mobile cardiac two-dimensional ultrasound is performed. Ultrasound can be performed within minutes as a radiology registrar is on call 24 hours a day in the trauma unit. Any visible fluid or clot is considered to be positive.

	Results

Top Abstract Introduction Material and methods Results Comment References

 
Perioperative findings
Group I
Mortality was 30% (7 of 23 patients). All deaths occurred in the 11 patients undergoing emergency room thoracotomy. Four deaths were early (2 patients exsanguinated and 2 patients had no return of cardiac rhythm); three late deaths were due to hypoxic brain damage (2 patients), and multiple organ failure. Six patients were intubated, resuscitated, and taken to the operating room (within a mean time of 26 minutes) with no mortality. Another 6 patients arrived with no recordable pulse or blood pressure and responded well to fluid resuscitation alone; 3 patients became completely stable, and 3 remained with a SBP more than 90 mm Hg before reaching the operating room.

Group II
Mortality was 9% (3 of 33 patients). One patient with a transected left anterior descendens artery died in the operating room, another had a right coronary artery injury (ligation was performed), and the third had postoperative bleeding.

The average amount of preoperative fluid given was only 1.1 L, compared with 2 L in each of the other three groups.

Group III
No deaths occurred.

Group IV
No deaths occurred. Fifteen patients in this group (34%) were transferred from a peripheral hospital. Some survived a trip as long as 5 hours by road. One patient initially refused consent and therefore, only came to operation after 10 hours.

Clinical findings
Precordial wound, left parasternal wound, initial hypotension, and distended neck veins were most common, in that order. Most patients (78%) presented with clinical tamponade, and 14% only with hemorrhage. Thirty-nine (90%) of group IV patients had either a globular heart on chest roentgenogram, CVP over 12 cm of water, or at least one clinical sign of tamponade. All these patients had a pulse less than 105 beats/min and SBP more than 100 mm Hg. Most had a SBP more than 120 mm Hg. Seven patients (16%) in group IV had no signs of tamponade or globular heart on chest roentgenogram, but only a CVP elevated between 10 and 20 cm of water, and a precordial wound.

Chest roentgenograms
About 80% of patients in groups III and IV had chest roentgenograms performed, 60% of these demonstrating a globular heart. All were supine films.

Pericardiocentesis
This was performed in 6 severely shocked patients, and improvement was noted in 4 patients.

Cardiac ultrasound
Cardiac ultrasound was performed in 5 patients in group II (15%), 14 in group III (48%), and 37 in group IV (86%). There were six false negatives; 3 of these patients had large hemothoraces and were taken to the operating room for bleeding, 2 patients deteriorated and were taken directly to the operating room, and 1 patient was diagnosed 5 days later by echocardiogram as the chest roentgenogram had an abnormal cardiac contour. There were no false positives.

Anatomic location
Six patients had through-and-through injuries (Table 1).

Clotted laceration
Clotted lacerations were documented in 4 of 23 group I patients (17%), no group II patients, 5 of 29 group III patients (17%), and 21 of 43 group IV patients (48%).

Observation of stable patients with a positive cardiac ultrasound
Three completely stable patients with a slightly elevated CVP and positive cardiac ultrasound, but no other clinical signs of tamponade, were observed, presuming that they had clotted lacerations. Two patients in this group suddenly collapsed later, needing emergency operation. One patient was discharged home.

Surgical approach
Sternotomy was performed in the majority of patients (100 patients). Left anterior thoracotomy was performed for emergency room thoracotomy, as well as in some stable patients. Combined approaches were used in 3 patients. There were no negative explorations.

Complications
Twenty-four patients (18.7%) required postoperative ventilation, and 15 (11%) needed prolonged ventilation for more than 24 hours. Reexploration for bleeding was needed in 3 patients (2.3%). One patient who underwent suture of a clotted left ventricular laceration developed a left-sided hemiparesis with incomplete resolution. Delayed complications included sinus of Valsalva fistula into the right ventricle needing repair, postintubation tracheal stricture (resection performed), and mitral valve defect (patient being followed up).

	Comment

Top Abstract Introduction Material and methods Results Comment References

 
The low mortality in this series, in keeping with a previous publication [3], is probably a reflection of fewer patients reaching the hospital alive [4], as well as aggressive resuscitation. It has been reported that only 10% of these patients reach the hospital alive [5, 6], but this figure is obviously higher in countries with better transfer systems—resulting in more severely injured patients being treated. In the United States, gunshot wounds are more common, whereas most of our patients had single or multiple stabs with smallish skin lacerations (usually < 2 cm). Associated injuries (such as the internal mammary artery) are common, and should be actively investigated.

All patients presenting with a precordial wound should be suspected of having a cardiac laceration. Sixty percent of patients with a parasternal or precordial stab have been confirmed to have a cardiac laceration [7], and 85% of patients confirmed to have a cardiac injury will have a stab in this area [8]. It is generally accepted that a patient with a precordial wound and unexplained elevation of the CVP should undergo subxiphoid window to exclude tamponade [1, 9].

Aggressive fluid resuscitation cannot be overemphasized. In the 1950s it was already observed that adequate volume loading could completely stabilize some patients, these patients being treated with pericardiocentesis alone [10, 11]. Rapid resuscitation has been shown to convert more than 85% of initially unstable patients from potential thoracotomy in the emergency room to candidates for exploration in the operating room—with greatly decreased mortality [1]. The fact that 17% of our patients in group I had clotted lacerations implies a more salvageable situation. We have been able to transfer some of these patients to the operating room after intubation, volume loading, and placement of tube thoracostomies. A CVP of more than 20 cm of water in these patients implies successful resuscitation [1].

Conservative treatment of 9 completely stable patients with a chest roentgenogram suggesting pneumopericardium has been successful [12]. With our high incidence of clotted lacerations in group IV patients, it was thought at one time to be safe to observe this group. However, 2 of 3 patients being observed deteriorated. Furthermore, the fact that most clotted lacerations bleed during repair has led us to decide that there is no place for observing these patients. There is enough evidence of later deterioration or death of these stable patients when the diagnosis has been missed [13]. In a recent study comparing mortality data with hospital admissions of a total of 1,198 patients, 70 (6%) reached the hospital alive with a confirmed diagnosis of a penetrating cardiac injury. However, seven missed injuries were found postmortem in patients who had been admitted and discharged from the hospital [14]. A missed injury may present with delayed tamponade [15].

Confirmation of diagnosis may be difficult. Our trauma unit’s aggressive approach to a patient with a precordial wound results in a larger subset of patients who do not have a cardiac laceration but have a persistently elevated CVP, with the concern that a negative exploration may be performed. Pericardiocentesis may have false negatives as high as 80% as a result of the presence of a clot [16], and false positives are also frequent [9, 17]. Some researchers [3, 8] agree that pericardiocentesis may have a therapeutic use in the severely shocked patient; other investigators [16] believe that the complication rate may be too high.

Subxiphoid pericardial window has been adopted with reasonable results [18], but a 20% false-negative rate has been seen [19]. True negative subxiphoid explorations may occur in 8% of patients [9]. Some researchers [16, 20] believe that it may waste valuable resuscitative time and allow catastrophic hemorrhage to occur. The surgeon will usually convert to a sternotomy, whereas a small anterior thoracotomy may be sufficient in a stable patient. In limited cases, however, it may allow adequate exposure for repair [18]. We performed only two subxiphoid explorations, and have abandoned this in favor of cardiac ultrasound.

Echocardiography has been used in the acute setting since 1984, when a pseudoaneurysm of the left ventricle following a stab was diagnosed [21]. With subsequent use it was not only found to be reliable, but allowed earlier decisions to be made [22]. A prospective study was made where 73 stable patients with precordial wounds underwent two-dimensional echocardiogram followed by subxiphoid window. There were 64 negative and 9 positive windows. Only 1 patient had a false-negative echocardiogram. Echocardiogram was found to be 96% accurate, 97% specific, and 90% sensitive, and it was concluded that it is the investigation of choice in the stable patient [23].

Liberal use of echocardiogram decreased the preoperative delay from 42.4 to 15.5 minutes, and increased survival from 57% to 100% [24]. The need for subxiphoid window has been eliminated by some surgeons, who proceed directly with sternotomy [7]. Another study with echocardiogram followed by subxiphoid window showed that echocardiogram was as accurate, sensitive, and specific in patients without hemothorax, but became less reliable in the presence of hemothorax [25].

Echocardiography may not always be immediately available, but with cardiac ultrasound we can rapidly confirm the diagnosis, decrease delays, and have an acceptable false-negative rate of 11%. A large clot in the pericardium is demonstrated with ease, as is a smaller fluid collection (Fig 1). Nonoperative management can thus be safely reserved for patients with normal vital signs and a negative cardiac ultrasound. In patients with a negative ultrasound, but clinically more likely to have tamponade, echocardiogram or subxiphoid window should be considered.

View larger version (116K): [in this window] [in a new window]   Fig 1. Cardiac ultrasound of a completely stable patient with a small central sternal laceration. Sternotomy was performed and a clotted laceration of the right ventricle was found, which started bleeding during suture.

	References

Top Abstract Introduction Material and methods Results Comment References

1. Buchman T.G., Phillips J., Menker J.B. Recognition, resuscitation and management of patients with penetrating cardiac injuries. Surg Gynecol Obstet 1992;174:205-210.[Medline]
2. Bolton J.W.R., Bynoe R.P., Lazar H.L., Almond C.H. Two-dimensional echocardiography in the evaluation of penetrating intrapericardial injuries. Ann Thorac Surg 1993;56:506-509.[Abstract]
3. Knott-Craig C.J., Dalton R.P., Rossouw G.J., Barnard P.M. Penetrating cardiac trauma. Ann Thorac Surg 1992;53:1006-1009.[Abstract]
4. Arreola-Risa C., Rhee P., Boyle E.M., Maier R.V., Jurkovich G.G., Foy H.M. Factors influencing outcome in stab wounds of the heart. Am J Surg 1995;169:553-555.[Medline]
5. Kulshrestha P., Iyer K.S., Das B., Balram A. A clinical and autopsy profile. J Trauma 1988;28:844-847.[Medline]
6. Oakland C., Vivian J. Penetrating cardiac injuries. Br Med J 1987;295:502.
7. Asensio J.A., Montgomery Stewart B., Murray B., et al. Penetrating cardiac injuries. Surg Clin North Am 1996;76:685-724.[Medline]
8. Robbs J.V., Baker L.W. Cardiovascular trauma. Curr Problems Surg 1984;XXI:12-25.
9. Arom K.V., Richardson J.D., Webb G., Grover F.L., Trinkle J.K. Subxiphoid pericardial window in patients with suspected traumatic pericardial tamponade. Ann Thorac Surg 1977;23:545-549.[Abstract]
10. Cooley D.A., Dunn J.R., Brockman H.L., DeBakey M.E. Treatment of penetrating wounds of the heart. Surgery 1955;37:882-889.
11. Blalock A., Ravitch M.M. A consideration of the non-operative treatment of cardiac tamponade resulting from wounds of the heart. Surgery 1943;14:157-162.
12. Michelow B.J., Bremner C.G. Penetrating cardiac injuries. J Trauma 1987;27:398-401.[Medline]
13. Mayor-Davies J.A., D’Egidio A., Schein M. "Missed stabbed hearts"—pitfalls in the diagnosis of penetrating cardiac injuries. Report of four cases. S Afr J Surg 1992;30:18-19.[Medline]
14. Campbell N.C., Thomson S.R., Muckart D.J., Meumann C.M., Van-Middelkoop I., Botha J.B. Review of 1198 cases of penetrating cardiac trauma. Br J Surg 1997;84:1737-1740.[Medline]
15. Aaland M.O., Sherman R.T. Delayed pericardial tamponade in penetrating chest trauma. J Trauma 1991;31:1563-1565.[Medline]
16. Demetriades D. Cardiac wounds. Experience with 70 patients. Ann Surg 1986;203:315-317.[Medline]
17. Trinkle J.K. Penetrating heart wounds. Ann Thorac Surg 1984;38:181-182.[Medline]
18. Johnson S.B., Nielson J.L., Sako E.Y., et al. Penetrating intrapericardial wounds. Ann Thorac Surg 1995;60:117-121.[Abstract/Free Full Text]
19. Attar S., Suter C.M., Hankins J.R., Sequeira A., McLaughlin J.S. Penetrating cardiac injuries. Ann Thorac Surg 1991;51:711-716.[Abstract]
20. Tavares S., Hankins J.R., Moulton A.L., et al. Management of penetrating cardiac injuries. Ann Thorac Surg 1984;38:183-186.[Abstract]
21. Choo M.H., Chia B.L., Chia F.K., et al. Penetrating cardiac injury evaluated by two-dimensional echocardiography. Am Heart J 1984;108:417-420.[Medline]
22. De Priest W., Barish R., Almquist T., et al. Echocardiographic diagnosis of acute pericardial effusion in penetrating chest trauma. Am J Emerg Med 1988;6:21-23.[Medline]
23. Jimenez E., Martin M., Krukenkamp I., Barret J. Subxiphoid pericardiotomy versus echocardiography. Surgery 1990;108:676-680.[Medline]
24. Plummer D., Bunette D., Asinger R., Ruiz E. Emergency department echocardiography improves outcome in penetrating cardiac injuries. Ann Emerg Med 1992;21:709-712.[Medline]
25. Meyer D., Jessen M., Grayburn P. Use of echocardiography to detect occult cardiac injury after penetrating thoracic trauma. J Trauma 1995;39:902-909.[Medline]

This article has been cited by other articles:

				A. N. Patel, C. Brennig, J. Cotner, M. A. Lovitt, M. L. Foreman, R. E. Wood, and H. C. Urschel Jr Successful diagnosis of penetrating cardiac injury using surgeon-performed sonography Ann. Thorac. Surg., December 1, 2003; 76(6): 2043 - 2047. [Abstract] [Full Text] [PDF]

				D. G. Harris, J. T. Janson, J. Van Wyk, J. Pretorius, and G. J. Rossouw Delayed pericardial effusion following stab wounds to the chest Eur. J. Cardiothorac. Surg., April 1, 2003; 23(4): 473 - 476. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Harris, D. G. Articles by Rossouw, G. J. Search for Related Content PubMed PubMed Citation Articles by Harris, D. G. Articles by Rossouw, G. J.