Hemodynamic Consequence of Interventional Cardiac Catheterization in the Early Postoperative Period after Congenital Heart Surgery

: While still considered a high-risk procedure, cardiac catheterization during the early postoperative period is being performed more frequently in the current era. Limited data are currently available concerning the acute hemodynamic consequences of these procedures. Therefore, the purpose of this study was to evaluate the safety/efﬁcacy of cardiac catheterization performed within thirty days of congenital heart surgery. We completed a retrospective review of all catheterizations within 30 days of congenital heart surgery. Procedures were performed due to failure to progress or hemodynamic deterioration. There were 1873 congenital heart surgeries during the study period. One hundred and three (6.2%) patients with a median age of 124 days underwent catheterization. Sixty-three cases received interventions, and forty patients underwent diagnostic catheterization. Early cardiac catheterization did not show a signiﬁcant immediate change in the hemodynamics or inotrope score. Survival for patients undergoing diagnostic Cath (81%) did not differ signiﬁcantly from the intervention group (89%). Although cardiac catheterization was performed on patients at the highest risk for death in the postoperative period, catheter intervention did not increase the risk of death. Those patients undergoing catheter intervention did not seem to experience major adverse events but achieved mild improvement in tissue perfusion.


Introduction
The evolution in the management of complex congenital heart disease has proven the usefulness of diagnostic and interventional catheterization [1][2][3]. In efforts to optimize the results, patient selection for catheterizations has previously been limited to non-acutely sick children no longer in the postoperative period. While still considered to be a high-risk procedure, cardiac catheterization in an acutely ill patient still within the early postoperative period has grown in popularity and applications over the past decades as a means to manage a complicated postoperative course [4,5]. The hemodynamic impact of these procedures on the immediate cardiac intensive care unit (CICU) course has yet to be defined in the literature.
For the past decade, it has been our practice to take critically ill patients to the Cath lab for evaluation and intervention if residual lesions are suspected, regardless of the time from surgery. Our working hypothesis has been that patients with lesions amenable to intervention will have significant improvements in the hemodynamic parameters, as well as inotrope requirements, shortly upon return to the CICU. With this in mind, we sought to quantify the changes in clinical hemodynamic variables such as heart rate, pulse oximetry, blood pressure, and inotrope requirement following the procedure to define the catheterization's immediate effects on the CICU course. Furthermore, in an effort to better define this subpopulation, the patient characteristics (diagnosis, surgical staging, clinical syndromes, indications for procedure, catheterization findings and interventions, time to catheterization, and reoperation rates) were reviewed and correlated with the days to discharge, mortality, and intervention versus non-interventional groupings.

Patient Selection
The inclusion criteria for this retrospective chart review incorporated all patients under twenty-one years of age who underwent catheterization within thirty days of congenital heart surgery at Nicklaus Children's Hospital (formerly known as Miami Children's Hospital) between January 2002 and February 2009. Patients whose clinical conditions did not warrant ICU admission or were electively scheduled for catheterization were excluded. A waiver for consent was granted by institutional review, given the retrospective approach of the study.

Catheterization Procedure
All catheterizations were performed by a pediatric interventional cardiologist in a bi-planar digital laboratory. General anesthesia was administered under the supervision of a pediatric cardiac anesthesiologist with the presence or immediate availability of a cardiac intensivist, a congenital heart surgeon, and supporting operating room staff for all cases. Standard techniques were followed for all balloon angioplasties, stents, valvuplasties, and for vascular occlusion with devices or particles. Previously published criteria were applied to determine the success or failure of interventional procedures [6-9].

Data Collection
Heart rate, systolic/diastolic blood pressure, and pulse oximetry were dependent variables considered valuable in defining the hemodynamic response to catheterization. Utilizing CICU charting records maintained in the EMTEK electronic nursing program, vital documentation was reviewed specifically for measurements taken immediately prior to leaving the unit, immediately upon arrival to the unit following the procedure, and twenty-four hours following the procedure. To define the effect of catheterization on the inotrope requirements, weight-based dosing rates for dopamine, epinephrine, and milrinone were recorded immediately prior to the procedure, upon return to the CICU, and again at twenty-four hours. An inotrope score was then calculated for each patient using a previously described scoring system [10]. Finally, in an effort to define the changes in renal function and tissue perfusion, the serum creatinine and lactate levels were reviewed and recorded if the specimen collection time corresponded to the pre-procedure, immediate post-procedure, and twenty-four-hour post-procedure times.

Statistical Analysis
The pre-and post-procedure variables were compared using paired t-tests or Wilcoxon rank-sum analysis as appropriate for normality utilizing Sigma Stat, version 3.1. Significance was considered at a p-value < 0.05.

Pre-Cath Patient Characteristics
During the study period, a total of 1873 congenital heart surgeries were performed, of which 1673 surgeries required cardiopulmonary bypass. Of the total surgical cases, 103 patients (6.2%) at a median age of 124 days (5 days-16.7 years; neonates n = 43, infants n = 37) and a mean weight of 8.1 kg underwent 116 catheterizations on the mean postoperative day 11.1. The diagnostic group underwent investigation at age 33 days compared to the interventional at 172 days (p = 0.005). Interventional patients were 1.4 times larger in weight than diagnostic patients (p = 0.001). Diagnostic patients were twice as likely to have had an open chest at any time during their admission (RR = 2.09, p = 0.009), and single ventricle physiology was more prevalent in the interventional group, though without statistical significance (73.9% versus 55.8%, p = 0.071) ( Table 1). Hypoplastic left heart syndrome, Tetralogy of Fallot/pulmonary atresia with VSD, and double outlet right ventricle anomalies accounted for 59% of the primary diagnoses (Table 2). Fontan was the predominant surgical stage at the time of catheterization (Table 3). Failure to wean inotropic and ventilatory support was the most common indication for catheterization (Table 4).

Catheterization Procedure
Sixty-three patients underwent 73 interventional catheterizations, during which 97 interventions were performed. Forty-eight percent of all procedures revealed new findings, with pulmonary artery stenosis (21.6%) and collateralization (12.9%) being the most common ( Table 5). Stenting of the pulmonary arteries, particle/device occlusion of aortopulmonary collaterals, and balloon angioplasty of pulmonary arteries accounted for 50.5%, 23.7%, and 16.5% of all interventions, respectively ( Table 6). The success rate for all interventions was 97%.

Complications
Among the entire cohort, twelve patients encountered a total of fourteen complications, with all but three occurring in the interventional group. Of the eleven complications within the interventional group, there were seven arrhythmias, two pulmonary artery intimal tears without hemorrhage, and one atrial septal stent device that migrated and required surgical retrieval. In the diagnostic complication group, there was one arrhythmia and two vascular site complications (Table 7).

Hemodynamic and Post-Procedure Characteristics
Prior to catheterization, there were no significant differences in heart rate, systolic, diastolic, pulse oximetry, serum lactate, inotrope score, and cardiopulmonary support use (CPS) between interventional and diagnostic patients; however, the diagnostic group was 19% more likely to be intubated than the interventional group (p = 0.057). At one hour post-catheterization, no significant changes were seen in any of the hemodynamic parameters, and at 24 h post-procedure, the only statistically significant change was in the serum lactate, albeit clinically insignificant (−0.3 mg/dL, p = 0.01) ( Table 8).

Comparison by Mortality
During the study period, there were a total of 1770 surgeries not requiring emergent Cath with a mortality prevalence of 5.8% and early and late mortality rates of 1.5% and 4.2%, respectively. The reoperation rates, median days to discharge following both Cath and surgery, and mortalities and time to mortality lacked significance between the two groups (Table 1). For the entire cohort, the thirty-day and late mortalities were 11.7% and 17.5%, respectively, for a total mortality prevalence of 30%. The interventional and diagnostic groups lacked statistical differences in the early mortality rates (7.9% and 17.5%, respectively), as well as late mortality (14.3% and 22.5%, respectively).
To further delineate the risk factors, the entire cohort was separated into early mortalities (<31 days after surgery), late deaths, and survivors. Early mortalities were more likely to be younger, smaller in weight, and to have received a Norwood procedure prior to the Cath. An open chest during Cath or any time during admission, as well as cardiopulmonary support before and after the Cath, was more prevalent among early deaths. One hundred percent of the early mortalities were intubated pre-and 24 h post-cath. There were no statistical differences in the fluoroscopy times and contrast administration during the Cath (Table 9).
In comparing late mortalities against the survivors, prior to the Cath, the late mortality group was more likely to also have recently undergone a Norwood procedure, be smaller in weight, intubated, hypoxic, and have a higher heart rate than the survival group. Interestingly, the survivor group was more likely to undergo Cath sooner, nine days postsurgery, than the mortality group at 16.5 days (p = 0.01). Following the Cath, the median 24-h inotrope scores were 50% lower in the survivors (p = 0.039). Cardiopulmonary support, pre-and post-lactates, fluoroscopy time, contrast dosing, and reoperation rates were not statistically different (Table 10).

Discussion
Originally, it was hypothesized that, by palliating residual lesions, interventional patients would see an immediate improvement in the hemodynamic status upon returning from the Cath lab compared to diagnostic patients. Our data failed to demonstrate improvement in the hemodynamic parameters such as heart rate, blood pressure, lactate, and oxygenation within 24 h of the procedure. The explanation for this lack of improvement may include no true benefit from the intervention or delay in the resolution of hemodynamic instability past the 24-h window that we investigated. Significant differences in the time to discharge and ultimate mortality outcomes are lacking, which raises the question as to whether these interventions are truly necessary. Due to ethical considerations, we are limited in our ability to characterize the benefits of interventions in the early postoperative period against true controls (patients with lesions but withheld from the intervention). A hypothesis that would prove difficult to test could be made that a true control group would have poorer cardiac output, exaggerated hemodynamic instability, longer hospitalizations, and higher mortality rates than our interventional group, all of which would justify exposing a critically ill patient to the risk of catheterization to palliate residual lesions.
Concern for the increased mortality risk of an invasive procedure in critically ill patients cannot be dismissed. While Asoh et al. encountered a six-month mortality rate approaching 50%, our series again demonstrated a low mortality rate at 6 months of 29% for all Caths and 22% for interventional catheterizations [11]. Furthermore, characteristics such as an open chest at Cath (17% and 22%), the interval from surgery to catheterization (10 and 7-10 days), and presenting congenital lesions are similar between the two study groups. Similar procedural success rates (95% versus 98%) suggest that the technical aspects of interventions can be reliably reproduced. However, variations in the reoperation rates, 37% in the Asoh series and 17% in ours, suggest that complete appreciation of the original and coexisting lesions, as well as an alternative surgical technique, can have a favorable impact on postoperative survival. Appreciation of concomitant lesions has improved with the application of three-dimensional imaging modalities such as angiographic computed tomography and traditional computed tomography, which can provide a definitive anatomical diagnosis of undiscovered lesions [12]. Recent studies have demonstrated growing trends in using three-dimensional rotational angiographies, which has been associated with superior outcomes in diagnostic and interventional procedures [13,14]. Hybrid procedures, initially intended for initial palliation of the hypoplastic left heart and complex single ventricle, have now been expanded to address septal defects and stenotic pulmonary arteries. Further employment of the hybrid operating room can alleviate the burden of postoperative lesions, thus reducing the length of ICU stays [15].
Taking a critically ill postoperative patient to the Cath lab requires considering the risks and benefits of the procedure, which may not be clear when the patient's ICU course has been tenuous; however, delaying a catheter examination also carries its own risk/benefit ratio. Asoh et al. described in their series of sixty-two postoperative procedures that a delayed catheter procedure was associated with higher in-house mortality, particularly when two to three weeks out from the index surgery [11]. We found a similar association with survivors in our series who were taken to the Cath lab on average a week sooner than the late mortality group. This pattern of an earlier invasive examination would suggest that earlier intervention and/or understanding of the hemodynamics can have a significant impact on ICU management and the subsequent outcomes.
Nicholson et al. described their experience in performing cardiac Cath in the early postoperative period and reported a low mortality rate and acceptable feasibility [16]. These results mirrored the current series, in which 97% of interventions were successful, with a thirty-day survival of 92%. Furthermore, a recent article showed that cardiac catheterizations within six weeks of cardiac surgery had a low rate of complications [17]. Our series encountered the associated weaknesses of a retrospective review within a single institution. With the aid of electronic charting, accurate charting of vitals and lab draws can be improved, but the variations cannot be completely eliminated.

Conclusions
Catheter investigation in the early postoperative period continues to play a reliable role in both clarifying labile hemodynamics, as well as addressing residual lesions with a variety of interventions. With careful patient selection, these interventions can be safely employed and spare the patient from return visits to the operating room. An earlier catheterization interrogation may lead to the earlier detection of unfavorable hemodynamics or lesions, which can improve the clinical course and morbidity. Informed Consent Statement: Waiver for consent was granted by institutional review, given the retrospective approach of the study. Data Availability Statement: Data available upon reasonable request from the corresponding author.

Conflicts of Interest:
The authors have no relevant financial or non-financial interest to disclose.