- Research article
- Open Access
Dying from cardiac tamponade
World Journal of Emergency Surgery volume 2, Article number: 22 (2007)
To determine the causes of cardiac tamponade (CT), focussing especially on haemopericardium (HP), as a terminal mode of death, within a 430,000 rural English population.
Our hospital mortuary register and, all postmortem reports between 1995 and 2004 inclusive, were interrogated for patients dying of CT or HP. The causes of CT/HP and selected morphological characteristics were then determined.
14,368 postmortems were performed in this period: of these, 461 patients died of CT. Three cases were due to non-haemorrhagic pericardial effusion. HP accounted for the remaining 458 cases of which, five were post-traumatic, 311 followed rupture of an acute myocardial infarction (RAMI), 138 after intra-pericardial rupture of dissecting ascending aortic aneurysms (RD3A) and four were due to miscellaneous causes.
HP was more commonly due to RAMI. Men tended to die from RAMI or RD3A earlier than women. RAMI or RD3A were commoner in men <70 yrs, but more frequent in women after this.
Two thirds of RAMI were associated with coronary artery thrombosis. Anterior free wall rupture was commonest overall, and in women, but posterior free wall rupture was commoner in men.
The volume of intrapericardial blood in RAMI (mean = 440 ml) and RD3A (mean = 498 ml) varied between 150 and 1000 ml: intrapericardial blood volume was greater in men than in women dying from either RAMI or RD3A.
At postmortem, CT is most often related to HP, attributable to either RAMI or intrapericardial RD3A. Post-traumatic and other causes of CT are infrequent.
The pathophysiology of cardiac tamponade (CT), as a cause of death, is related to an increase in intrapericardial fluid pressure that exceeds atrial venous pressures, thereby impeding venous return to the heart. Rapidly evolving HP (200 to 300 ml) is more likely to cause death from CT than slowly evolving pericardial fluid accumulation (500 to 2000 ml), the latter allowing for accommodation of greater volumes due to gradual distension of the pericardial sac. The normal volume of pericardial fluid (30 to 50 ml) reflects a balance between production and reabsorption.
The causes of CT include active or passive pericardial effusion and haemopericardium (HP) consequent on trauma, iatrogenic intervention or, either rupture of an acute myocardial infarction (RAMI) or intrapericardial rupture of a dissecting ascending aortic aneurysm (RD3A)[1, 2]. HP commonly follows RAMI, RD3A or trauma [1–3]; it has also been described in association with malignancy[3, 4], chemotherapy, homicide, pacing wire or central venous catheter[8, 9] insertion, endocardial biopsy, open heart surgery, interventional coronary artery procedures (atheroablative, angioplastic or stenting), exercise stress test, electroconvulsive therapy, coronary artery vasculitis or dissecting aneurysm, myocardial abscess, infective endocarditis and, during amniocentesis or in the prenatal period. CT been described in association with pneumopericardium  and as a complication of intrapericardial drain insertion. HP has been described as a complication of 5% to 10% of patients with AMI .
Post-AMI myocardial rupture includes ventricular free wall rupture, ventricular septal rupture or papillary muscle rupture [1, 3]. Risk factors for this include age >60 yrs, female gender, pre-existing hypertension and lack of left ventricular wall hypertrophy.
The lateral wall, at mid-ventricular level, is said to be the most common site for post-infarction free-wall rupture. However, the lateral and inferior aspects of the left ventricle have been reported as equally susceptible to post-infarctive rupture .
We have reviewed all cases of patients dying of cardiac tamponade in our hospital over a ten-year period to determine the causes of cardiac tamponade and highlighted some associated morphological features.
Our department performs postmortems at the request of Her Majesty (HM) Coroner, in the County of Cornwall, that has a population of ~430,000. Almost all deaths in this catchment area, falling within the jurisdiction of HM Coroner, are sent to our mortuary for a postmortem. These deaths are principally those within the community, or hospital, that fall within the purview of HM Coroner : for the most part, these postmortems are performed because there is no firmly established cause of death.
The Mortuary Death Register of the Department Histopathology, Royal Cornwall Hospital, was reviewed for causes of death listed as either "haemopericardium" or "cardiac tamponade" between 1995 and 2004 (inclusive). These results were compared to a search of our textual postmortem report database using "moper" (a unique string of letters in the word "haemopericardium") and "tamponade" as separate search criteria within the text of the reports; the results of these were amalgamated and duplicates discarded.
All postmortem reports were retrieved and the cause of death confirmed as being attributable to HP or CT; the following parameters were then confirmed and retrieved from each postmortem:
1. The gender and age of patients
2. The causes of the CT or HP
3. If HP due to myocardial infarction, the
a. Presence of coronary artery thrombosis
b. Site of myocardial rupture and
4. Volume of intrapericardial haemorrhage
14,368 postmortems were performed at the Royal Cornwall Hospital in the ten years under review. There were 461 cases of CT (3.2% of all postmortems), of which there were three cases of non-haemorrhagic effusion (passive = 2; infective = 1), five cases of post-traumatic HP and four miscellaneous causes (Table 1); these were excluded from subsequent analysis. Of the rest (n = 449; 98%), CT was due HP (Figure 1) secondary to RAMI (n = 311; 69%) or RD3A (n = 138; 31%) (Figure 2).
Death due to HP, as a consequence of either RAMI or RD3A, is commoner in women than men (Figure 3). RAMI is a more common cause of HP than RD3A in both men and women.
The age range of non-traumatic HP in men (33 to 99 years) and women (21 to 99 years) are similar. The mean age of men dying of HP/CT as a result of RAMI (73.4 ± 8.6) (n = 120) and RD3A (72.5 ± 13.0) (n = 62) is less than those of women dying of HP/CT from RAMI (78.7 ± 9.0) (n = 191) and RD3A (76.6 ± 11.1) (n = 76). The incidence of RAMI and RD3A in each age group is shown in Figures 4 &5; death from HP due to either cause is commoner in men than female before 70 yr but reverses after this and is commoner in women.
RAMI was responsible for most causes of death (69%; n = 311) due to HP, with an overall male (n = 120; 39%) to female ratio (n = 191; 61%) of 1:1.6.
Coronary artery thrombosis (CAT) was recognised in 210 postmortems (67%: 82 M, 128 F); CAT was not identified in the remainder (n = 101: 38 M, 63 F). There were no significant differences in the ages of men or women with CAT as a cause of RAMI.
The site of ventricular rupture was defined in all postmortem reports (table 2), with the anterior wall (33.8%) being the most common site of rupture. Cumulative figures show anterior (42.8%) and posterior ventricular wall rupture (37.3%) being more common than apical (6%) or lateral (22.5%) ventricular wall rupture. These figures are probably skewed by the larger numbers of females in the RAMI group.
Men were more likely to have posterior (12.2%) than anterior (10.6%) or lateral (7.4%) ventricular wall rupture; women, by contrast, had anterior (23.2%) ventricular wall rupture more commonly than posterior (19.3%) or lateral (5.8%) wall rupture.
RD3A, HP and CT
Patients dying from intrapericardial RD3A formed a smaller but significant group (30.5%; n = 138) of patients dying from HP. This mode of death was slightly more frequent in women (n = 76; 55%) than in men (n = 62; 45%) (Figure 3).
Intrapericardial blood volume
The volume of blood in the pericardial space was recorded in 246 (54%) instances (99 M, 147 F); this ranged between 150 ml and 1000 ml of blood in RAMI or RD3A (figure 6).
In addition, 800 ml of blood was seen in association with pericardial angiosarcoma, 750 ml in association with metastatic lung carcinoma and 1650 ml in association with malignant pericarditis. The volume of blood in association with Warfarin therapy was not recorded.
A difference in intrapericardial blood volume was noted between men and women whether as a consequence of either RAMI (M = 473 ml, F = 418 ml) or RD3A (M = 514 ml, F = 474 ml).
Death was associated with a slightly smaller volume of HP in-patients dying as a consequence of RAMI (mean = 440 ml) when compared to RD3A (mean = 498 ml).
There is little evidence in the literature reflecting on the causes or demographics of HP in routine postmortem practice. With the exception of individual case reports , most of the evidence available in standard textbooks is non-referenced.
Only 3.2% of all our deaths, over a 10-year period, were attributable to CT. In our series RAMI and RD3A were the principal causes of CT, causing death as a consequence of HP. Pericardial effusion, post-traumatic CT and miscellaneous causes of CT together formed only a small group of CT-related deaths.
Although the lateral wall at mid-ventricular level  or the inferior wall are said to be the most common sites for post-infarction free-wall rupture , in our series anterior left ventricular wall rupture was more common than rupture at other ventricular wall sites. This might be a reflection of the greater numbers of women in our study, anterior free wall rupture being more common in women and posterior free wall rupture in men.
By contrast to traditional risk factors of female gender and age >60 years, men appear to die from CT due either RAMI or RD3A earlier than women. Indeed, death from either of these causes is commoner in men than women before, but not after, 70 years of age. This probably reflects the effects of cardiogenic risks in women evolving in the postmenopausal period and of men dying earlier from other causes, including AMI without myocardial rupture.
Traditionally, acute cardiac tamponade is associated with between 200 ml and 300 ml of sudden accumulation of intrapericardial fluid or, in chronic slowly evolving accumulation, of volumes between 1000 and 2000 ml[1, 3]. This contrasts our experience with volumes of between 440 ml and 500 ml in association with CT due to acute pericardial accumulation of blood, in women and men respectively. In cases where HP volumes were larger than this, we speculate that this was associated with a slow leak, with or without a terminal phase of catastrophic rapid haemorrhage. We also demonstrated that slightly less intrapericardial blood volume is associated with RAMI than RD3A in fatal HP.
In conclusion we have defined the aetiology of CT in a postmortem population in rural South-West England and have described some of the salient morphological features associated with CT.
Acute Myocardial Infarction
Ruptured Acute Myocardial Infarction
Dissecting Ascending Aortic Aneurysm
Rupture Dissecting Ascending Aortic Aneurysm
Schoen FJ: The heart. Robbins and Cotran Pathologic Basis of Disease 7th edition. Edited by: Kumar V, Abbas AK, Fausto N. Philadelphia, Elsevier Saunders; 1999:555-618.
Schoen FJ: Blood vessels. Robbins and Cotran Pathologic Basis of Disease. Edited by: Kumar V, Abbas AK, Fausto N. 1999, Philadelphia , Elsevier Saunders, 511-554.
Anderson JR, Hunt I: Cardiac Surgery. Bailey and Love's Short Practice of Surgery 24th edition. Edited by: Russell RCG, Williams NS, Bulstrode CJK. London , Arnold: Hodder Headline Group; 2004:917.
Haskell RJ, French WJ: Cardiac tamponade as the initial presentation of malignancy. Chest. 1985, 70-73.
Angelucci E, Mariotti E, Lucarelli G, Baronciani D, Cesaroni P, Durazzi SM, Galimberti M, Giardini C, Muretto P, Polchi P: Sudden cardiac tamponade after chemotherapy for marrow transplantation in thalassaemia. Lancet. 1992, 339 (8788): 287-289. 10.1016/0140-6736(92)91343-7.
Altun G, Altun A, Yilmaz A: Hemopericardium-related fatalities: a 10-year medicolegal autopsy experience. Cardiology. 2005, 104 (3): 133-137. 10.1159/000087633.
Howell C, Bergin JD: A case report of pacemaker lead perforation causing late pericardial effusion and subacute cardiac tamponade. J Cardiovasc Nurs. 2005, 20 (4): 271-275.
Greenspoon JS, Masaki DI, Kurz CR: Cardiac tamponade in pregnancy during central hyperalimentation. Obstet Gynecol. 1989, 70: 465-466.
Collier PE, Goodman GB, Kurz CR: Cardiac tamponade caused by central venous catheter perforation of the heart: a preventablecomplication. J Am Coll Surg. 1995, 181: 459-463.
Craven CM, Allred T, Garry SL, Pickrell J, Buys SS: Three cases of fatal cardiac tamponade following ventricular endocardial biopsy. Arch Pathol Lab Med. 1990, 114 (8): 836-839.
Kuvin JT, Harati NA, Pandian NG, Bojar RM, Khabbaz KR: Postoperative cardiac tamponade in the modern surgical era. Ann Thorac Surg. 2002, 74 (4): 1148-1153. 10.1016/S0003-4975(02)03837-7.
Fejka M, Dixon SR, Safian RD, O'Neill WW, Grines CL, Finta B, Marcovitz PA, Kahn JK: Diagnosis, management, and clinical outcome of cardiac tamponade complicating percutaneous coronary intervention. Am J Cardiol. 2002, 90 (11): 1183-1186. 10.1016/S0002-9149(02)02831-X.
Weinberg L, Kandasamy K, Evans SJ, Mathew J: Fatal cardiac rupture during stress exercise testing: case series and review of the literature. South Med J. 2003, 96 (11): 1151-1153. 10.1097/01.SMJ.0000055036.73825.E2.
Ali PB, Tidmarsh MD: Cardiac rupture during electroconvulsive therapy. Anaesthesia. 1997, 52 (9): 884-886. 10.1111/j.1365-2044.1997.193-az0330.x.
Hunsaker DM, Hunsaker JC, Adams KC, Noonan JA, Ackermann DM: Fatal Kawasaki disease due to coronary aneurysm rupture with massive cardiac tamponade. J Ky Med Assoc. 2003, 101 (6): 233-238.
Cheung A, Chan CW: Dissecting aneurysm of coronary artery presenting as cardiac tamponade. N Z Med J. 1990, 103 (886): 129-130.
Fan CC, Andersen BR, Sahgal S: Isolated myocardial abscess causing coronary artery rupture and fatal hemopericardium. Arch Pathol Lab Med. 1994, 118 (10): 1023-1025.
Berner HW, Seisler EP, Barlow J: Fetal cardiac tamponade. A complication of amniocentesis. Obstet Gynecol. 1972, 40 (4): 599-604.
Wolf PL, Jones KL, Longway SR, Benirschke K, Bloor C: Prenatal death from acute myocardial infarction and cardiac tamponade due to embolus from the placenta. Am Heart J. 1985, 109 (3 Pt 1): 603-605.
Stechmiller J, Seymour SF, McGinn CJ: Cardiac tamponade resulting from pneumopericardium: case report and implications for the critical care nurse. Heart Lung. 1987, 16 (4): 442-448.
Oliva PB, Hammill SC, Edwards WD: Cardiac rupture, a clinically predictable complication of acute myocardial infarction: report of 70 cases with clinicopathologic correlations. J Am Coll Cardiol. 1993, 22 (3): 720-726.
Biedrzyck O, Baithun S: Seasonal variation in mortality from myocardial infarction and haemopericardium. A postmortem study. J Clin Pathol. 2006, 59: 64-66. 10.1136/jcp.2005.025965.
Cerek B, Shah PK: Acute coronary syndromes: complications of acute myocardial infarction. Cardiology. 2003, Elsevier, 373-Edited by Crawford MH, DiMarco JP, Paulus WJ, 2nd edition
Marquand P: The Coroner's inquest. Introduction to medical law. 2001, London , Arnold, 60-66.
The author(s) declare that they have no competing interests.
Authors’ original submitted files for images
About this article
Cite this article
Swaminathan, A., Kandaswamy, K., Powari, M. et al. Dying from cardiac tamponade. World J Emerg Surg 2, 22 (2007) doi:10.1186/1749-7922-2-22
- Infective Endocarditis
- Pericardial Effusion
- Cardiac Tamponade
- Wall Rupture
- Ventricular Septal Rupture