|
|
||||||||
Ann Thorac Surg 2003;76:152-156
© 2003 The Society of Thoracic Surgeons
a Division of Cardiothoracic Surgery, Childrens Hospital, Omaha, Nebraska, USA
b Division of Cardiothoracic Surgery, Childrens Hospital, Pittsburgh, Pennsylvania, USA
Accepted for publication January 17, 2003.
* Address reprint requests to Dr Fenton, Cardiothoracic Surgery, Childrens Hospital, 8200 Dodge St, Omaha, NE 68114, USA.
e-mail: kfenton{at}chsomaha.org
| Abstract |
|---|
|
|
|---|
METHODS: From January 1991 through December 2000, a total of 146 infants aged 60 days or less underwent placement of systemic-topulmonary artery shunts and were discharged from the hospital alive. Inpatient, outpatient, and autopsy records were reviewed.
RESULTS: Indications for surgery were single-ventricle anatomy in 90 cases and complex double-ventricle anatomy in 56. Of the patients, 21 (14%) died after discharge and before further planned surgery. Of these 21 infants, 17 (81%) were clinically doing well before sudden death. Autopsies were obtained in 15 cases and attributed the cause of death to shunt thrombosis in 5 infants (33%), myocardial infarction in 2 (13%), and pneumonia or lung disease in 3. Five autopsies were nondiagnostic. The mortality of patients discharged on aspirin (11.1%) was almost identical to that of patients discharged on no anticoagulation (12.3%). Four infants with sudden death had been notably irritable for 24 to 48 hours before death.
CONCLUSIONS: There is a significant incidence of sudden death among infants who have undergone shunting. Death may be preceded by unexplained irritability, and such symptoms should therefore be carefully evaluated. Autopsy-proven shunt thrombosis is one of the leading causes of interim sudden death, and aspirin therapy may not be helpful. Options to reduce interim mortality include alternative regimens of anticoagulation (such as lowmolecular weight heparin), alternative conduit material, and earlier reoperation.
| Introduction |
|---|
|
|
|---|
Because perioperative mortality for neonatal palliation of hypoplastic left heart syndrome (HLHS) has fallen [1, 2], it has become increasingly apparent that there is a significant rate of interim death (ie, death after discharge from the shunt procedure and before second stage palliation or definitive repair). These deaths are particularly striking because they commonly occur in infants who are at home and, to all appearances, thriving. Recent studies [1, 3, 4] have reported an interim mortality in patients with HLHS ranging from 4% to 15%. The purpose of the current investigation was to study the risk and cause of interim death for all infants with systemic-to-pulmonary artery shunts. We hypothesized the following: (1) that all infants with aortopulmonary shunts, independent of anatomy, are at risk for sudden death; (2) that death is related in some cases to occlusion of the shunt; and (3) that anticoagulation with aspirin decreases the risk of shunt thrombosis.
| Patients and methods |
|---|
|
|
|---|
The study was reviewed and approved by the institutional review board of the Childrens Hospital (Pittsburgh, PA). Inpatient and outpatient medical records were reviewed for the 146 hospital survivors. In cases in which an autopsy was performed, the report was obtained, reviewed, and correlated with clinical information.
Data were retrospectively collected. Variables examined included cardiac and noncardiac diagnoses, age and weight at operation, concomitant procedures performed, shunt material and size, and use of any anticoagulant or antithrombotic agents after hospital discharge. Outcome variables included death, shunt thrombosis, and subsequent planned operation. Statistics were performed for categorical variables using contingency table analysis.
| Results |
|---|
|
|
|---|
|
|
Interim death (Table 2) was unrelated to aspirin administration (p = 0.49), shunt size (p = 0.67), or the presence of another source of pulmonary blood flow in addition to the shunt (p = 0.40). There was only a trend toward a higher rate of interim death in infants with single-ventricle anatomy (p = 0.10). Interim death occurred at a median age of 68 days (mean 136 days, Fig 2).
|
|
There were 10 thrombosed shunts in the 169 patients who underwent operation, including 2 inpatients who died. Eight thromboses occurred in 3.5-mm shunts and two in 4-mm shunts. There was thus a higher incidence of shunt thrombosis in smaller shunts (13% for shunts 3 to 3.5 mm vs 2.4% for shunts 4 to 5mm, p < 0.02). There was a higher incidence in patients on aspirin (12% vs 2.3% without aspirin, p < 0.02), but patients with smaller shunts were more likely to be placed on aspirin (p < 0.02). Overall mortality was almost identical with and without aspirin (Table 2). Two of the patients with shunt thrombosis had been recently evaluated for emesis.
| Comment |
|---|
|
|
|---|
Previous studies have focused attention on the interim mortality of patients with HLHS. Interim death occurred in 4% to 15% of patients with HLHS in three recent studies [1, 3, 4], and in 24% of HLHS patients in our series. Residual cardiac lesions have been identified on autopsy in only a small number of patients [1]. In one multicenter study interim death was related to low institutional case volume, older age at operation, and smaller ascending aorta or mitral valve [4]. It may be possible to prevent interim death by using a home surveillance program to help select patients who should undergo a second-stage procedure at an earlier date [3]. Interim death in shunted patients without HLHS has not been previously investigated.
Because of the physiologic consequences of the shunt itself, we suspected that any infant with a shunt would have a risk of interim mortality. These data demonstrate our hypothesis to be correct; perhaps the significant (7.1%, Fig 1) incidence of interim death in patients with double-ventricle anatomy is more surprising and more concerning than is the higher percentage in patients with HLHS. This is an important factor to keep in mind, particularly when making the surgical plans for patients whose anatomy may permit an alternative surgical option.
We can categorize the causes of interim death in our patients based on shunt physiology. Two patients in our series died of inadequate systemic cardiac output related to excessive pulmonary blood flow. Two with documented myocardial infarction had no evident coronary problems at autopsy, and thus most likely had limited coronary blood flow as a result of insufficient systemic cardiac output or low diastolic blood pressure. Inadequate systemic perfusion is also the most likely cause of the multiorgan failure in 1 patient. Five patients died suddenly with no clinical indication of the cause of death and had nondiagnostic autopsies. The possibility that these infants died of an arrhythmia related to low cardiac output and myocardial ischemia is supported by the presence of soft tissue edema at autopsy in 2 of the 5. A prior ventricular incision, a potential source for arrhythmia, was present in only 1 patient who had undergone excision of a left ventricular diverticulum at the time of his shunt. Thus, as many as 10 of the 21 interim deaths (48%) may be attributed to insufficient system cardiac output or inadequate coronary perfusion. This is supported by the fact that 13 of 14 patients with recorded autopsy weights were at or below the 25th percentile. To minimize the chance of interim death, it will be important to optimize systemic cardiac output and diastolic blood pressure to maintain coronary perfusion in infants who have undergone shunt procedures. Other current work emphasizes the importance of balancing systemic and pulmonary blood flow [5, 6]. Future investigation of the optimal values for titration of afterload reducing agents may be warranted.
An additional and related consideration is the limited ability of these infants to withstand any physiologic stress. Of our 21 patients, 3 died after a minor medical procedure, 2 of whom had documented myocardial ischemia at autopsy. Four other infants died 24 to 48 hours after being evaluated (1 by telephone, 3 in person) for mild illnesses. Of these patients, 2 had nondiagnostic autopsies and 2 had shunt thromboses. In the infants with nondiagnostic autopsies, we speculate that they were unable to tolerate the stress of fever and mild illness. It is also possible, however, that their symptoms (mostly fussiness or irritability) were related to ischemia. Other investigators have also noted a relationship of nonspecific complaints with sudden death [3]. Certainly close monitoring of infants with systemic-to-pulmonary artery shunts is required at the time of any physiologic stress. Judicious use of inotropic support may be warranted in individual cases.
Shunt thrombosis was identified in 5 autopsied patients, and was statistically associated with smaller shunt size and with aspirin use. Other studies have also reported low rates of shunt thrombosis [79]. Our data fail to demonstrate any beneficial effect of aspirin to prevent shunt thrombosis. The positive relationship between aspirin use and shunt thrombosis is related to the fact that aspirin was intentionally more commonly prescribed for patients with smaller shunts. Death resulting from shunt thrombosis occurred only in patients with single-ventricle anatomy. Two patients (40%) who died of shunt thrombosis had been evaluated with acute nonspecific symptoms within 48 hours of sudden death; both had reportedly been "vomiting" or "spitting up." We speculate that even mild dehydration may increase the risk of shunt thrombosis. Shunt thrombosis is a fatal event in patients with single-ventricle anatomy. When patients who survived a thrombosed shunt are included, there is no association between shunt thrombosis and single-ventricle anatomy (Fig 1). It is possible that an alternative regimen of anticoagulation or a different conduit material may favorably affect the incidence of shunt thrombosis [9].
The optimal conduit size for systemic-to-pulmonary artery shunts is a subject of long-standing debate [5, 6]. It is likely that the trend in recent years toward using smaller shunts has contributed to a decrease in perioperative mortality due to excessive pulmonary blood flow but perhaps also to an increase in the incidence of shunt thrombosis. Discussion of shunt size is made more difficult by variations in technique such as the location of the proximal end of the shunt and its length. Certainly any technical problems with the shunt contribute to the chance of its thrombosis; this did not appear to be the case in any of our autopsied patients. Unfortunately, the same patients who are poor candidates for large shunts because of low cardiac reserve are also theoretically at higher risk for shunt thrombosis. It is these patients on whom we must focus our efforts to find either an effective regimen of anticoagulation or, more likely, an alternative conduit material.
We speculate that the new ventricular-to-pulmonary artery conduits may also have a lower rate of thrombosis. A shunt placed between the right ventricle and the pulmonary artery in patients with HLHS allows adequate pulmonary blood flow while preserving diastolic blood pressure and thus limiting the risk of coronary steal. The larger conduit in this location may also decrease the risk of thrombosis. In one preliminary study using this technique [10], however, there were no early deaths but interim mortality was 11%, all from shunt thrombosis. Finally, stenting of the ductus arteriosus and banding of the pulmonary arteries has been presented as yet another first-stage alternative, with a reported rate of interim death of 9% in one small series [11].
Examination of the timing of interim death is instructive (Fig 2). Outpatient death occurred at mean age 136 days (range 39 to 382 days). Consideration must be given to planning the subsequent operation as early as is feasible, and proceeding with the operation as early as planned (ie, avoidance of "scheduling" delays). Attention should be directed specifically at proceeding before the development of symptoms such as desaturation.
Pertinent negatives in our study include an absence of any cases of endocarditis causing death, an absence of any autopsy findings of aspiration, and an absence of residual anatomic lesions contributing to death. We often invoke aspiration as a possible cause of sudden deterioration or death in infants; however, we may need to rethink this based on these autopsy results. The absence of residual lesions probably reflects some combination of growth in experience and patient selection (patients with residua die in hospital), but subtle residual lesions may have been missed.
Limitations of our study include its retrospective nature, the somewhat erratic use of aspirin, the wide variety of anatomic subtypes, the exclusion of inpatient deaths, and the absence of autopsies in a minority of patients. Our data demonstrate that aspirin use is associated with smaller shunt size; in fact, both are related to later date of operation and to an intentional change in practice on the part of individual surgeons. We were unable to demonstrate that antiplatelet therapy prevents shunt thrombosis. This should not be taken as an implication that aspirin is not needed or not helpful but, rather, as a caution that it may not be sufficient. We have intentionally elected to include all patients with shunts to demostrate that all are at risk for sudden outpatient death, even though this admittedly makes it difficult to compare patients. It is most surprising that there is such a high rate of interim death in patients with double-ventricle anatomy and in those with systemic left ventricles. Inpatient deaths were intentionally omitted from the analysis because, in most cases, they seemed to be causally different. In fact, there was one inpatient death in an infant who was on the ward ready to be discharged, who perhaps causally belongs to the outpatient group. She had a nondiagnostic autopsy. There were also two interim deaths in patients who died in the intensive care unit of chronic problems who perhaps would fit better in the inpatient death group. We elected to limit the study to patients who died after initial hospital discharge because this seems to best separate the groups.
In conclusion, careful attention to the physiology of infants with systemic-to-pulmonary artery shunts can allow us to minimize interim mortality. We can begin with the obvious: avoidance of shunts when appropriate, and proceeding to second stage palliation as early as possible. Systemic cardiac output and coronary perfusion must be optimized at base line, and must be carefully preserved and monitored in times of physiologic stress. Dehydration must be assiduously avoided to minimize the chance of shunt thrombosis, and careful attention must be directed toward any infant with a shunt who has an apparently mild illness or irritability. Any child with a systemic-to-pulmonary artery shunt and any clinical symptoms (such as irritability, emesis, or poor appetite) should be promptly evaluated with echocardiography or cardiac catheterization. The infant must be closely observed, and consideration must be given to earlier second operation.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
K.-J. Lee, A. Hinek, R. R. Chaturvedi, C. L. Almeida, O. Honjo, G. Koren, and L. N. Benson Rapamycin-Eluting Stents in the Arterial Duct: Experimental Observations in the Pig Model Circulation, April 21, 2009; 119(15): 2078 - 2085. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Mohammadi, O. Benhameid, A. Campbell, J. Potts, J. Joza, H. Al-Habib, S. Sett, and J. Le Blanc Could we still improve early and interim outcome after prosthetic systemic-pulmonary shunt? A risk factors analysis Eur. J. Cardiothorac. Surg., September 1, 2008; 34(3): 545 - 549. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. A. Hehir, T. E. Dominguez, J. A. Ballweg, C. Ravishankar, B. S. Marino, G. L. Bird, S. C. Nicolson, T. L. Spray, J. W. Gaynor, and S. Tabbutt Risk factors for interstage death after stage 1 reconstruction of hypoplastic left heart syndrome and variants J. Thorac. Cardiovasc. Surg., July 1, 2008; 136(1): 94 - 99. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. M. Bradley, C. C. Erdem, T.-Y. Hsia, A. M. Atz, V. Bandisode, and J. M. Ringewald Right Ventricle-to-Pulmonary Artery Shunt: Alternative Palliation in Infants With Inadequate Pulmonary Blood Flow Prior to Two-Ventricle Repair Ann. Thorac. Surg., July 1, 2008; 86(1): 183 - 188. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Monagle, E. Chalmers, A. Chan, G. deVeber, F. Kirkham, P. Massicotte, and A. D. Michelson Antithrombotic Therapy in Neonates and Children: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition) Chest, June 1, 2008; 133(6_suppl): 887S - 968S. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. McAlvin, M. L. Clabby, P. M. Kirshbom, K. R. Kanter, B. E. Kogon, and W. T. Mahle Routine Immunizations and Adverse Events in Infants With Single-Ventricle Physiology Ann. Thorac. Surg., October 1, 2007; 84(4): 1316 - 1319. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. M. Cholette, J. S. Rubenstein, G. M. Alfieris, M. P. McDermott, W. G. Harmon, R. Vermilion, M. P. Eaton, J. J. Gangemi, and N. B. Lerner Elevated Risk of Thrombosis in Neonates Undergoing Initial Palliative Cardiac Surgery Ann. Thorac. Surg., October 1, 2007; 84(4): 1320 - 1325. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. N. Fenton, K. Lessman, K. Glogowski, S. Fogg, and K. F. Duncan Cerebral Oxygen Saturation Does Not Normalize Until After Stage 2 Single Ventricle Palliation Ann. Thorac. Surg., April 1, 2007; 83(4): 1431 - 1436. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. M. Graham, G. A. Forbus, S. M. Bradley, G. S. Shirali, and A. M. Atz Incidence and outcome of cardiopulmonary resuscitation in patients with shunted single ventricle: Advantage of right ventricle to pulmonary artery shunt J. Thorac. Cardiovasc. Surg., May 1, 2006; 131(5): e7 - e8. [Full Text] [PDF] |
||||
![]() |
C. D. Myers, J. H. Boyd, R. G. Presson Jr, P. Vijay, A. C. Coats, J. W. Brown, and M. D. Rodefeld Neonatal Cavopulmonary Assist: Pulsatile Versus Steady-Flow Pulmonary Perfusion Ann. Thorac. Surg., January 1, 2006; 81(1): 257 - 263. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. D. Myers, K. Mattix, R. G. Presson Jr, P. Vijay, D. Maynes, K. N. Litwak, J. W. Brown, and M. D. Rodefeld Twenty-Four Hour Cardiopulmonary Stability in a Model of Assisted Newborn Fontan Circulation Ann. Thorac. Surg., January 1, 2006; 81(1): 264 - 271. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. L. Skinner, L. A. Halstead, C. S. Rubinstein, A. M. Atz, D. Andrews, and S. M. Bradley Laryngopharyngeal dysfunction after the Norwood procedure J. Thorac. Cardiovasc. Surg., November 1, 2005; 130(5): 1293 - 1301. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. J. Wells, R. J. Yu, A. S. Batra, H. Monforte, C. Sintek, and V. A. Starnes Obstruction in Modified Blalock Shunts: A Quantitative Analysis With Clinical Correlation Ann. Thorac. Surg., June 1, 2005; 79(6): 2072 - 2076. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Kirjavainen, S. Viskari, O. Pitkanen, and E. Jokinen Infants with univentricular heart have reduced heart rate and blood pressure responses to side motion and altered responses to head-up tilt J Appl Physiol, February 1, 2005; 98(2): 518 - 525. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Pizarro, T. Mroczek, E. Malec, and W. I. Norwood Right Ventricle to Pulmonary Artery Conduit Reduces Interim Mortality After Stage 1 Norwood for Hypoplastic Left Heart Syndrome Ann. Thorac. Surg., December 1, 2004; 78(6): 1959 - 1964. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. N. Fenton, F. A. Pigula, S. K. Gandhi, L. Russo, and K. F. Duncan Interim Mortality in Pulmonary Atresia With Intact Ventricular Septum Ann. Thorac. Surg., December 1, 2004; 78(6): 1994 - 1998. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. M. Bradley, J. M. Simsic, T. C. McQuinn, D. M. Habib, G. S. Shirali, and A. M. Atz Hemodynamic status after the Norwood procedure: A comparison of right ventricle-to-pulmonary artery connection versus modified blalock-taussig shunt Ann. Thorac. Surg., September 1, 2004; 78(3): 933 - 941. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Sittiwangkul, A. Azakie, G. S. Van Arsdell, W. G. Williams, and B. W. McCrindle Outcomes of tricuspid atresia in the Fontan era Ann. Thorac. Surg., March 1, 2004; 77(3): 889 - 894. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. D. Rodefeld, J. H. Boyd, C. D. Myers, R. G. Presson Jr, W. W. Wagner Jr, and J. W. Brown Cavopulmonary assist in the neonate: an alternative strategy for single-ventricle palliation J. Thorac. Cardiovasc. Surg., March 1, 2004; 127(3): 705 - 711. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |