Introduction

Today’s medical environment is marked by financial constraints. There is increasing pressure to optimize not only medical care but also resource utilization. This notion has also affected the way we approach the treatment of children requiring surgery for congenital heart disease (CHD). For example, reducing intensive care unit (ICU) and hospital length of stay (LOS) without compromising patient’s safety translates into much‐needed cost savings.

Although the overarching goal of providing efficient and safe care has not changed, not only the terminology but also the approach to achieving this goal has seen a significant evolution. The following chapter will focus on perioperative management, including early tracheal extubation, adequate postoperative pain control, rapid mobilization, and hospital discharge.

Background and history

The focus on not only improving surgical techniques but also on providing efficient care to patients undergoing cardiac surgery was introduced in the mid‐1980s. Diagnosis‐related groups were implemented nationally into Medicare reimbursement in the United States in 1983. To slow down the rapidly increasing costs of providing medical care, this reimbursement model incentivized efficient resource use, and a shorter length of ICU and hospital stay was suddenly linked to financial benefits for the hospitals. At the same time, in congenital heart surgery (CHS), extubation at the end of the surgical procedure was almost a necessity, as reliable mechanical ventilators and sedatives with minimal side‐effects for small patients were not widely available.

In 1980, Barash et al. published their experience with early extubation in 197 patients aged less than 3 years including neonates; of them, 61% were successfully extubated in the operating room (OR) [1]. The authors noted that “in this era of cost containment, any technique that allows maximal use of resources without jeopardizing patient safety is welcome.” The concept of the day of admission surgery, early‐extubation, ‐mobilization, and ‐hospital discharge became popular and eventually routine practice in adults undergoing cardiac surgery in the late 1990s. This was especially true when crowded ICUs slowed the expansion of cardiac surgery programs attempting to gain a share of the increasing number of coronary revascularizations being performed [2]. In 1994, Engelman et al. first coined the term “fast‐track” in their paper, describing a complete care plan for patients undergoing cardiac surgery, which was associated with reduced hospital LOS [3].

In the field of CHS, surgical techniques developed rapidly and palliation and repair of CHD in even younger and previously inoperable children became feasible. Reports about favorable effects of high‐dose opioid‐based anesthesia emerged in the early 1990s, promising to reduce stress response, morbidity, and mortality [4, 5]. Administering high doses of opioids subsequently became the standard anesthetic technique in children undergoing increasingly complex CHS. This opioid‐based anesthetic technique, however, necessitated prolonged postoperative mechanical ventilation (MV), and early tracheal extubation was no longer an option. Subsequently, shorter‐acting drugs and anesthetic agents with less pronounced hemodynamic effects were introduced and the benefit and need of a high‐dose opioid‐based anesthetic were challenged [6, 7]. More recently, bundled care payment models have been implemented for specific procedures, and healthcare providers are once more encouraged to efficiently manage costs. At the same time, not meeting quality of care measures is linked to financial penalties.

Many of these past and recent changes have led hospital administrators, as well as individual healthcare providers, to re‐think and investigate established, yet often inefficient, patient management strategies.

Enhanced recovery after pediatric cardiac surgery

In line with these efforts, the term “enhanced recovery after surgery” or ERAS, originally coined by Henrik Kehlet for colorectal surgery in 1997 [8], was adopted by many surgical specialties, including cardiac and pediatric cardiac surgery [9]. This patient management strategy, paraphrased by the acronym ERACS (enhanced recovery after cardiac surgery), focuses on a patient‐centered, evidence‐based, multidisciplinary approach to cardiac, including pediatric cardiac, surgery. ERACS strategy includes many if not most of the elements of fast‐tracking children and adults undergoing surgery for CHD, that is, multidisciplinary approach, day of admission surgery, normothermia at the end of the procedure, multimodal pain management strategy, early ‐extubation, ‐mobilization, ‐enteral feeding, ‐ICU and –hospital discharge (Figure 24.1). The goals are also very similar, i.e., providing efficient, patient‐centered patient care, that would help accelerate functional recovery and minimize surgical stress, thereby reducing the LOS, improving patient outcomes, optimizing resource use, and reducing associated costs. It further emphasizes routine monitoring and reporting of quality metrics and patient outcomes, aimed at continuously improving the operational aspects of this approach. The American Association for Thoracic Surgery published a consensus document on enhanced recovery after pediatric cardiac surgery [10]. Key strategies that were identified for successfully implementing and maintaining an ERACS pathway include a multi‐disciplinary approach, appropriate patient selection, staff and patient/family education, protocol‐ and guideline‐driven practice. The task force concluded that ERACS does not increase morbidity and mortality after pediatric cardiac surgery. Patient and family satisfaction with perioperative care may be improved. Roy and colleagues described their initial experience after introducing ERACS to their CHS program [11]. Compared to a matched historical control group, ERACS implementation resulted in a slightly reduced time on mechanical ventilation and LOS in ICU, although, no difference in hospital LOS was noted.

Feasibility of fast‐tracking in congenital heart surgery

The multidisciplinary approach behind fast‐tracking CHS patients typically includes an anesthetic technique that facilitates early tracheal extubation. Early extubation is not consistently defined, and the term is frequently used when the endotracheal tube is removed within 6 hours after surgery [12, 13], but spans from immediate extubation in the OR up to 24 hours after surgery [10].

The feasibility and safety of extubating children immediately or soon after complex CHS had already been demonstrated in the early days of CHS, mostly out of necessity [1]. Schuller et al. reported their experience with early extubation in 1984 [14]. In 209 consecutive children undergoing complex open‐heart surgery, 88% of those older than 12 months were extubated in the OR. In the same year, Heard et al. published an article on early extubation following CHS in 220 patients; of whom 147 (67%) were extubated in the OR or within 6 hours of entering the ICU [15]. None of the patients required reintubation. This was accomplished even though many of the anesthetic agents used at that time are obsolete in today’s practice.

Despite the increasing popularity of a high‐opioid anesthetic technique following these early reports, individual centers continued to pursue an early extubation strategy. The published reports and the recent changes in healthcare practice including ERAS pathways encouraged many institutions to adopt fast‐tracking for at least some of their patients. At the time of this writing, the feasibility of fast‐tracking including early extubation in children undergoing CHS has been demonstrated for most patients undergoing CHS. This includes neonates and young children [16, 17], complex surgeries [18, 19], children with single‐ventricle physiology [20, 21], patients with significant co‐morbidities, [22] and even in pediatric heart transplantation [23]. In a recent multicenter international survey spanning 144 pediatric cardiac surgery centers in 29 countries, extubation in the OR at the conclusion of surgery was practiced by 76% of the responding anesthesiologists, mostly for lower surgical risk procedures [24]. Table 24.1 summarizes the current literature on the feasibility of fast‐tracking and early extubation in pediatric patients undergoing CHS.

Patient selection

When it comes to planning fast‐tracking including early extubation for a specific patient or case, it seems prudent to understand the current evidence and have some guidance regarding who is a good candidate for such an approach. Some preoperative factors are almost universally viewed as contraindications for early extubation after CHS. For example, very few patients who have an endotracheal tube in place before the procedure will be good candidates for such a strategy. However, in many institutions, this practice is further limited to selected patients and is based on the anesthesiologist’s or surgeon’s preferences. While some centers consider early extubation only above a certain age and for simple procedures, others expand such a strategy to the majority of their patients. In many, if not all, of the published reports, specific exclusion criteria were applied, and it is not clear if patients who were not included could have been extubated early as well. As a matter of fact, there are reports showing successful early extubation in even the most complex procedures. Therefore, when interpreting study findings, it is important to notice not only the study design, inclusion and exclusion criteria, but also if extubation was attempted or deferred. A simple association between perioperative parameters and early extubation alone, without the specifics of extubation attempt, will not yield reliable data beyond individual practice or preferences.

The most consistent findings associated with deferred extubation are higher procedure complexity, younger age, longer aortic cross‐clamp and cardiopulmonary bypass (CPB) times, and high‐dose inotrope use [30, 35–42]. This has also been confirmed by prospective studies. Kin et al. found higher procedure complexity classified with the risk adjustment for congenital heart surgery (RACHS) score [43], surgery involving aortic cross‐clamping, young age, and trisomy 21 to be independent predictors of the decision to not attempt extubation at the end of the procedure [32]. Very few patients in that series failed extubation once attempted. At first glance, these findings seem to classify selection criteria in a very simplified scheme. All listed factors are known preoperatively, and the findings seem to indicate that intraoperative factors such as bleeding, cardiac dysfunction requiring high inotropic support, generalized edema, etc., many of which are frequently encountered, are not significantly contributing to predicting early extubation. However, procedure complexity seems to capture many of these factors. Complex surgeries are frequently associated with longer aortic cross‐clamp and CPB times, more pronounced inflammatory response, and associated findings such as a higher incidence of bleeding, generalized edema, and higher inotropic requirements, which prohibit early removal of the endotracheal tube all by themselves [29, 44, 45]. In a prospective observational study, Garg et al. [34] collected information on OR extubation and reasons for deferring extubation to a later time point in 1,000 pediatric patients undergoing surgery for CHD within an 8‐month time period. Overall, 87.1% of patients were extubated in the OR, resulting in significantly shorter ICU LOS compared to a historical control group. Factors associated with deferred extubation included poor oxygenation, open sternum, ongoing bleeding, high inotrope use, and other reasons necessitating further observation in the ICU.

Aside from complex procedures, young age is one of the most frequently listed reasons for not considering early extubation [32]. In retrospective analyses, young age is consistently associated with not being fast‐tracked. Davis et al., for example, assessed factors associated with early extubation following CHS. Children older than 6 months were much more likely to be extubated early compared with younger patients [29]. Winch et al. [46] found that children younger than 3 months were less likely to be extubated early. However, extubation practice varies significantly and even neonates have been extubated successfully immediately after CHS [17]. Age also needs to be assessed in the context of the planned procedure. Certain more complex procedures may be performed later in life, while a simple procedure in an infant may pose a much smaller risk for not qualifying for early extubation. The fact that age alone is not a good selection criterion for fast‐tracking is also supported by the published literature. Kin et al. [32], for example, found that the relationship between age and attempted OR extubation was inconsistent and not linear. Beyond the age of 2 years, however, age did not impact the likelihood of OR extubation. Similarly, in Garg’s study [34] representing a large number of patients and a high overall extubation rate (87.1%), OR extubation was accomplished in 40% of neonates, 69.7% of 1–3‐month‐olds, 85.5% of 3 months–1‐year‐olds, and leveled off thereafter. In general, these data indicate that successful extubation in the OR can be accomplished more reliably in children older than 2–3 months of age. Younger infants and neonates need not be excluded from such a strategy; however, the incidence of encountering contraindications to early extubation at the end of the procedure or soon after in the ICU is typically much higher at this very young age.

Chromosomal abnormalities also account for patients not being considered for early extubation. While not a contraindication per se, patients with a genetic variation known to be associated with difficult airway instrumentation, or an increased risk of airway obstruction (e.g., trisomy 21), must be evaluated carefully before attempting early extubation. This is even more of a concern following a long CPB time, which is frequently associated with generalized edema possibly involving the upper airway.

Pre‐existing or worsening of pulmonary arterial hypertension (PAH) deserves particular attention. Many practitioners would not attempt early extubation in patients with pre‐existing PAH. However, this practice has been challenged, and there are reports of successful extubation following CHS even in patients with significant PAH [22]. In a retrospective analysis of factors associated with early extubation, PAH was not found to be an independent predictor of OR extubation [30], and preoperative PAH did not predict patients who were not extubated in the OR in a similar prospective study [32]. In the latter study, however, significant PAH following CPB was the most common reason for deferring extubation at the end of the procedure. While there is no question why PAH increases the risk for many adverse perioperative events [47–50], it is important to recognize that merely using the current definition of PAH (right‐sided pressures ≥20 mmHg) [51] does not accurately reflect such risks in children with CHD [52]. It is often difficult to predict the degree of PAH following surgical repair and, even more importantly, right ventricular failure merely by preoperative right‐sided pressures. Assessment of pulmonary vascular resistance (PVR), reactivity to pulmonary vasodilators, as well as the relationship between right‐sided pressures and systemic arterial pressure is typically included in the evaluation of PAH in children undergoing CHS [53]. Following weaning off CPB, patients can be reassessed, and even in the setting of pre‐existing PAH, patients can often be extubated as long as the right ventricular function is adequate, and no inhalational pulmonary vasodilators are required.

The likelihood of early extubation following CHS can be better predicted by combining several of the risk factors. Kin et al. developed a risk score model based on factors found to be independently associated with extubation in the OR following surgery [32]. All patients undergoing CPB surgery without aortic cross‐clamp and simple procedures (RACHS 1, for example, secundum atrial septal defect) were extubated in the OR. Higher RACHS scores and the addition of trisomy 21 as an additional risk factor increased the chance of deferring extubation. Similarly, Davis et al. found age <6 months, prematurity, congestive heart failure, and pulmonary hypertension to be associated with the chance of failing extubation within 24 hours after surgery. Adding individual factors increased the chance of failing early extubation [29]. Factors associated with early extubation in individual studies can be found in Table 24.1.

Implementation and maintaining fast‐tracking

The cumulative experience of fast‐tracking patients undergoing pediatric cardiac surgery clearly highlights the multi‐disciplinary approach required for the successful implementation of such a strategy. Many institutions where fast‐tracking has been practiced successfully have started with low complexity cases and older children, and eventually adopted this approach to a broader patient population [54]. With the rebranding of fast‐tracking and adoption of ERACS protocols, more data will likely emerge on implementing and maintaining this strategy in pediatric cardiac surgery also. Roy et al. [11] described their initial experience with introducing an enhanced recovery after CHS program at their institution. Key elements for implementation included an initial assessment of the evidence based ERACS interventions in pediatric CHS, multidisciplinary consensus building and development of institutional ERACS guidelines, as well as monthly sharing of outcome data for continuous process improvement. Despite this, the authors comment on the suboptimal adherence to ERACS guidelines.

The Pediatric Heart Network Collaborative Learning Study used collaborative learning principles to implement early extubation guidelines across four active study sites and five control sites [55]. Overall compliance with early extubation guideline adoption was high. An increase in early extubation rates, from 11.7% before to 66.9% after guideline implementation, were seen for patients undergoing tetralogy of Fallot (TOF) repair. No change was noted for coarctation of aorta repair. There was no increase in the reintubation rate during the study period. Although all active sites reported a significant decrease in time to extubation, significant variability between sites was noted (i.e., magnitude of change: −73.3% to −99.2%) [56]. Following this initial rollout, the investigators assessed if this initiative also impacted extubation practice in patients not targeted in the initial trial; “spillover” [57]. One year after the initial practice guideline was implemented, one of the four active study sites also noted a significantly higher extubation rate in patients undergoing ventricular septal defect repair. No change was seen in patients undergoing higher complexity surgeries. The same group of investigators also looked at the sustainability of the observed change in early extubation rate in patients undergoing TOF [58]. One year after initial study completion, the early extubation rate declined from 67% to 30%, and the median time to first extubation attempt increased from 4.5 to 13.5 hours. Only one out of four hospitals maintained the extubation rate achieved during the initial study. Sustaining and even expanding an early extubation practice, without doubt, require continuous work from all stakeholders involved, including transparency and reporting of outcomes and resource use.

Anesthesia technique

Although certain predictors can help select patients who are good candidates for early extubation even preoperatively, the ultimate decision to attempt extubation will always be deferred to the end of the procedure or ICU. For the purpose of fast‐tracking and, in particular, early extubation, an ideal anesthetic would allow early extubation and at the same time provide stable hemodynamics, adequate analgesia, and sufficient blunting of the stress response. In the early days of CHS, the use of potent inhalational anesthetics allowed for early extubation at the end of, or within a few hours after surgery. However, a mainly inhalational anesthetic‐based technique is not always tolerated and can be associated with significant side effects at higher concentrations. An opioid‐based anesthetic provides stable hemodynamics and superior suppression of stress response, but typically requires prolonged MV. The addition of a neuraxial anesthetic is advocated by some practitioners and will be discussed in more detail later in this chapter. There are concerns, however, about adequate evidence regarding the safety of a neuraxial technique in the cardiac surgery setting.

Made possible in large part by the introduction of improved and new anesthetic agents such as modern inhalational anesthetics, short‐acting opioids, hypnotics, and sedatives with favorable pharmacodynamics and pharmacokinetic profiles, all the above‐mentioned goals can be accomplished, and the side‐effects of each individual technique minimized. Such a balanced anesthetic consisting of an inhalation‐based technique supplemented with a short‐acting opioid is the basis of most of the modern anesthetic techniques aiming for fast‐tracking CHS patients and will be discussed in more detail here.

Patients who qualify for fast‐tracking are often admitted on the day of surgery, and frequently intravenous (IV) access has not yet been established. Therefore, an inhalational induction technique with sevoflurane is typically chosen for induction and is well tolerated in most patients. If true contraindications to an inhalational induction technique apply, IV access should be established, and ketamine or etomidate can be used as induction agents. With a true inhalation‐based fast‐tracking technique, the inhalational agent is administered throughout the case, including on CPB. The CPB machine must therefore be equipped with a vaporizer. Alternatively, a continuous infusion of IV agents with hypnotic properties can be used during CPB. As a short‐acting hypnotic with favorable pharmacokinetic and pharmacodynamics properties, propofol is frequently used during CPB for patients who are fast‐tracked [59]. If, however, a decrease in systemic vascular resistance (SVR) must be avoided, a high‐dose opioid technique or the use of ketamine should be considered. Ketamine administered as a continuous infusion maintains stable hemodynamics in most CHS patients.

Regarding IV opioid use, shorter‐acting opioids such as fentanyl are preferred. Early extubation can still be accomplished as long as the cumulative dose is limited, and larger repetitive doses avoided toward the end of the procedure. Frequently, not more than 5–10 μg/kg fentanyl is given for the whole case [25, 28, 30, 31], which is significantly less than the doses used for a high‐dose opioid technique. Remifentanil is an ultra‐short‐acting opioid that is metabolized by non‐specific esterases in plasma and tissues and minimally affected by CPB [60]. The use of remifentanil allows conduction of a high‐dose opioid anesthetic with the benefits of reducing stress response, but with a fast recovery facilitating early extubation [61–63]. The challenge is to provide adequate analgesia after an infusion of remifentanil is discontinued due to its rapid elimination. A neuraxial technique, non‐opioid analgesics, and dexmedetomidine can be helpful for this purpose.

Dexmedetomidine is a sedative agent that confers sedation by selectively binding to central α₂ adrenoceptors. Respiratory indices and upper airway patency are maintained during dexmedetomidine sedation in children [64–67]. In addition to sedative and analgesic‐sparing effects, dexmedetomidine use has also been reported to reduce stress response in patients undergoing cardiac surgery [68]. Additional favorable properties supporting dexmedetomidine’s role in the fast‐tracking setting are the reduced incidence of emergence delirium after inhalational anesthesia in children [69–71] and its possible role in reducing the incidence and even treatment of arrhythmias [72]. Additionally, emerging data mostly from in vitro studies show potential neuroprotective effects [73, 74]. Because of these favorable properties, dexmedetomidine is even used throughout the case by some practitioners and allows the reduction of opioid requirements for fast‐tracking purposes. Undesirable hemodynamic effects of dexmedetomidine such as bradycardia, hypotension, hypertension, decrease in cardiac output [75], as well as an increase in PVR, especially with high doses or fast bolus administration, have all been described. These side effects, however, are highly dose‐ and age‐dependent. Typically, dexmedetomidine is well tolerated in older children [76]; however, little data are available regarding safe doses in infants and particularly in neonates. There are data showing that dexmedetomidine clearance in neonates and infants is reduced [77–80], and accumulating drug levels causing bradycardic side effects may occur. In a recent pharmacokinetics and safety trial in neonates undergoing cardiac surgery, clearance was significantly reduced; however, dose adjustments resulted in a low incidence and severity of adverse events noted [81].

It must also be noted that at this point dexmedetomidine is still approved only by the US Food and Drug Administration (FDA) for use in adults for ICU sedation up to 24 hours. In clinical practice, however, dexmedetomidine has been widely used in children even for prolonged periods of time [82]. A Pediatric Postmarketing Pharmacovigilance and Drug Utilization Review can be found on the FDA web page [83].

In a Pediatric Heart Network‐sponsored multicenter Collaborative Learning Study, the change of anesthesia practice was compared before and after early extubation study guideline implementation [84]. Clinical practice guideline implementation aimed at early extubation in patients following repair of coarctation of the aorta or TOF was accompanied by a significant decrease in opioid and midazolam use, no change in intraoperative median volatile anesthetic use, and intraoperative dexmedetomidine administration was an independent predictor of early extubation.

In summary, fast‐tracking and early extubation are facilitated by modern anesthetic agents combined with strategies to provide good analgesia without significant respiratory depression.

Surgery and CPB considerations

Surgical and CPB techniques contribute significantly to the ability to fast‐track patients and have an impact on anesthetic management. Aside from the quality of the surgical repair, it is increasingly recognized that outcomes can be improved by reducing the inflammatory response [85] and limiting surgical trauma [86]. For example, avoiding or limiting prolonged periods of deep hypothermia [87] and circulatory arrest [88] reduces CPB duration and associated complications such as cardiac dysfunction, coagulopathy, and transfusion requirements [89]. Limiting the inflammatory response results in less generalized edema and improves pulmonary function [90–92], all important components to successfully fast‐tracking CHS patients. Minimally invasive surgical techniques [93–95], hybrid procedures [96], or alternate surgical access for selected CHD [97] can further contribute to limiting surgical trauma.

Significant changes have also been made to CPB and perfusion techniques. These improvements often benefit patients who are considered for fast‐tracking and include miniaturized CPB circuits [98–100], retrograde autologous priming [101], beating heart surgery techniques [102], less cooling [103–105], pulsatile CPB [106], and selective perfusion strategies [107]. Modified ultrafiltration (MUF) is also frequently part of a fast‐tracking protocol and has been shown to decrease edema and improve pulmonary and cardiac function [108–110]. Immediate effects on hematocrit and coagulation also help fast‐tracking [111]. Many of these techniques are controversial and not universally practiced, however, and the safety and possible outcome benefits are still being evaluated. Nevertheless, many of the currently ongoing changes to the practice of CHS promise to be beneficial to fast‐tracking.