Patients with multiple types of CHD may be palliated with a Fontan procedure. The Fontan procedure is suitable for patients who are not amenable to a two-ventricle repair due to either a small or absent ventricle, or due to complex anatomy which makes septation difficult. Reference Gewillig and Brown1–Reference Jolley, Colan, Rhodes and DiNardo3 In a recent publication from the Pediatric Heart Network, Fontan patients were found to have a 95% transplant-free survival at an average of 7 years of follow-up. Reference Atz, Zak and Mahony4 Many patients with Fontan physiology will live for decades Reference Ono, Boethig and Goerler5,Reference Reddy, Siehr Handler, Wu, Rabinovitch and Wright6 and thus may require diagnostic imaging and surgical procedures. Importantly, many of these interventions will require general anaesthesia or sedation.
Despite the large numbers of patients with Fontan circulations, analysis of the periprocedural outcomes following non-cardiac procedures has been limited. Our objectives are to describe the periprocedural course for non-cardiac procedures for patients with Fontan physiology and identify the predictors for discharge home the same day.
Methods
Institutional review board approval was obtained (IRB-P00022312). Administrative databases were interrogated for the interval between 1 January, 2013 and 31 December, 2019 to identify all patients who had undergone a previous Fontan procedure (n = 871). Anesthesia databases were then interrogated to identify any anaesthetic encounters for the cohort of Fontan patients. Anesthetics for cardiac surgery, cardiac catheterisation, electrophysiology or arrhythmia procedures, and cardiac imaging were all excluded. Anesthetics were excluded if they occurred either prior to the Fontan procedure (i.e. at the time of bidirectional Glenn) or if the patient no longer had a Fontan circulation following heart transplant, conversion to biventricular circulation, or had undergone a Fontan take-down to another single-ventricle palliation. An additional four anaesthetics were removed, because no medications were administered by an anaesthesiologist. This resulted in a final cohort of 182 patients who underwent 344 non-cardiac procedures under the care of a paediatric anaesthesiologist.
Data were collected using chart review and electronic capture of data from the electronic medical record to include basic demographics, underlying cardiac disease, type of Fontan, ventricular function and presence of valvular dysfunction on most recent echocardiogram, and baseline oxygen saturation. Details recorded from the anaesthesia record included the start time, the length of the anaesthetic, type of anaesthesia (general versus monitored anaesthesia care), airway management, use of neuromuscular blocking agents, use of inotropes, and the volumes of crystalloid and packed red blood cells administered. Comorbid conditions were assessed using a chronic condition indicator. Chronic condition indicator is an 18-category tool developed to classify International Classification of Diseases (ICD)-10 codes as chronic within administrative databases (Healthcare Cost and Utilization Project – HCUP). 7 A chronic condition is defined as one which is present for 12 months or greater, requires ongoing medical care, or results in a limitation of self-functioning. Patients were assigned 1 point for each system other than cardiovascular which was affected by 1 or more chronic conditions. The final disposition of the patients was determined based on the final location of recovery.
Procedures were categorised as diagnostic imaging, minor surgical procedures, or major surgical procedures. Minor procedures were defined as those which could reasonably be expected to be done with minimal sedation in a healthy, cooperative adults such as dental procedures, Botox injections, peripherally inserted central catheters, chest tubes, and liver biopsies. Also included in this category were direct laryngoscopy or endoscopy and dental rehabilitation. Major procedures included procedures with the potential for bleeding or major haemodynamic consequences such as most orthopaedic, neurosurgical, and general surgical procedures. Major procedures would be expected to require general anaesthesia.
Cardiopulmonary adverse events in patients included arrhythmias or tachy/bradycardias requiring intervention, increased oxygen requirement postoperatively to achieve baseline oxygen saturations after transfer to the recovery room, change to a more invasive airway (natural airway to laryngeal mask or endotracheal tube), cardiopulmonary resuscitation, or need for extracorporeal membrane oxygenation. These complications were assessed for the first 24 hours after anaesthesia and were only assessed for patients who did not recover in an ICU.
Statistics
Continuous data are reported as means and standard deviations, and categorical data are presented as frequencies and percentages. Continuous variables were compared using the Wilcoxon rank sum test, and categorical variables were analysed using the Chi-square test or Fisher’s exact test. A two-tailed p < 0.05 was considered statistically significant. Multivariable logistic regression analysis results are presented by adjusted odds ratios with 95% confidence intervals and p values. A two-tailed alpha level of 0.05 was used to determine statistical significance. Stata (version 16.1, StataCorp LLC., College Station, Texas) was used for all statistical computing.
Results
We identified 182 patients who had 344 non-cardiac procedures between 1 January, 2013 and 31 December, 2019. The median number of anaesthetics per patients was 2 (IQR 1–5). The median age was 11 years (IQR 5.2–18.0), the median weight was 31.5 kg (18–54.6), and 194 (56.4%) were male. The ASA category was II in 14 (4.1%), III in 221 (64.2%), IV in 108 (31.4%), and V in 1 (0.3%) and were emergent procedures in 39 (11.3%).
Patient underlying diagnoses included hypoplastic left heart syndrome (n = 134, 39.0%), tricuspid atresia (n = 59, 17.2%), double-outlet right ventricle (n = 39, 11.3%), pulmonary atresia (n = 35, 10.2%), double-inlet left ventricle (n = 22, 6.4%), unbalanced complete atrioventricular canal (n = 17, 4.9%), and other diagnoses (n = 38, 11.0%). It was a systemic left ventricle in 177 patients (51.5%). Two hundred and sixty-seven (77.6%) patients had a lateral tunnel or intracardiac Fontan and 77 (22.4%) had an extracardiac Fontan. No patients in this series had a classical Fontan. Three hundred twenty-one patients (93.3%) had a preoperative echocardiogram within the prior year. Twenty-seven patients (8.4%) had a significant lesion burden as determined by echocardiography and defined as severe atrioventricular valve regurgitation (n = 8) and/or qualitatively severe dysfunction of the systemic ventricle (n = 20).
The median value for the chronic condition indicator was 3 (IQR 0–5). The most common non-cardiac diagnoses were in the digestive system (39.8%). Of these, the 2 most common coded diagnoses were gastroesophageal reflux disease and gastrostomy. Psychological conditions were present in 38.4% including most frequently acute stress reaction and generalised anxiety disorder. Respiratory conditions were present in 23.8% with the most common diagnoses being asthma, obstructive sleep apnoea, and oxygen dependency.
Anesthetic and surgical information are included in Table 1. General anaesthesia was used in 289 (84.0%) of anaesthetic encounters. Induction techniques included inhalational in 87 (25.3%), intravenous 247 (n = 71.8%), or a combination in 10 (2.9%). Airway management was a tracheostomy in 3 (0.9%), an endotracheal tube in 215 (62.5%), a laryngeal mask or mask in 33 (9.6%), or a natural airway in 93 (27.0%) of anaesthetics. Invasive monitoring included an arterial line in 48 (14.0%) anaesthetics and a central line or peripherally inserted central catheter in 44 (12.8%). Inotropes/vasopressors were used in 134 anaesthetics (39.0%) and included dopamine (n = 110, 32.0%), ephedrine (n = 36, 10.5%), epinephrine (n = 3, 0.9%), and calcium gluconate (n = 24, 7.0%). Neuromuscular blockade was used in 164 anaesthetics (47.8%). Patients received a median of 7.2 mL/kg/hour of crystalloid (IQR 3.7–10.2 mL/kg/hour). Packed red blood cells were administered in 16 anaesthetics (4.7%).
Table 1. Anesthesia and surgery details.
There was a total of 20 regional blocks in 16 patients undergoing general anaesthesia. This included a transversus abdominis plane or rectus sheath block (n = 8), supraclavicular or infraclavicular (n = 3), lower limb (femoral, facia iliaca, and adductor canal) (n = 4), lumbar plexus (n = 3), paravertebral (n = 1), and epidural (n = 1).
In 274 cases (79.7%), the patients were transferred to a post-anesthesia care unit. The median length of stay in the post-anesthesia care unit was 98 minutes (IQR 67–157). The median day admitted for all patients postoperatively was 1 day (IQR 0–7). One hundred twenty-five (36.3%) were discharged on the same day, 156 (45.3%) were admitted for a median of 2 days (IQR 1–11 days), and 63 (18.3%) went to the ICU. All patients who were discharged home the same day had a follow-up phone call by our nursing staff. There were no readmissions within 24 hours after discharge.
There were 61 patients who initially recovered in an ICU postoperatively as scheduled and 2 cases patients initially admitted to the post-anesthesia care unit with a subsequent need for transfer to an ICU due to lethargy with hypotension. Cardiorespiratory adverse events occurred during an additional 16 anaesthetics; 2 were cardiac (1 tachyarrhythmias and 1 bradyarrhythmia) and 14 were respiratory (new oxygen requirements above baseline in the recovery room (n = 9), unanticipated changes in airway management intraoperatively (n = 3), laryngospasm (n = 1), and postoperative pleural effusion (n = 1)). There were no instances of cardiac arrest, respiratory arrest, extracorporeal membrane oxygenation initiation, or death within 24 hours. There were no complications related to regional anaesthesia in this cohort.
Univariate predictors of discharge on the same day are listed in Table 2. On univariate analysis, the following were associated with discharge: female gender (p < 0.001), ASA II or III (p < 0.001), elective procedures (p < 0.001), baseline oxygen saturations > 90% (p < 0.001), lower chronic condition index (p < 0.001), preoperative admission (p < 0.001), location of procedure (p < 0.001), anaesthesia duration (p < 0.001), anaesthesia start time (p < 0.001), type of procedure (p < 0.001), general anaesthesia (p = 0.033), type of airway device (p = 0.019), absence of arterial line (p < 0.001), absence of central venous access (p < 0.001), no inhaled anaesthetics used (p = 0.027), induction technique (p < 0.001), absence of inotropes used (p < 0.001), volume of crystalloid given per kg per hour (p < 0.001), and no blood transfusion (p < 0.001). On multivariable analysis, independent predictors that reduced the odds of same-day discharge included the chronic condition index (OR 0.91 per additional chronic condition, 95% CI 0.76–0.98, p = 0.022), undergoing a major surgical procedure (OR 0.17, 95% CI 0.05–0.64, p = 0.009), the use of intraoperative inotropes (OR 0.48, 95% CI 0.25–0.94, p = 0.031), and preoperative admission (OR = 0.24, 95% CI: 0.1–0.57, p = 0.001) (Table 3).
Table 2. Univariate associations with same-day discharge.
Continuous data are presented as median (IQR) and categorical data are presented as n (%).
Continuous variables were compared using the Wilcoxon rank sum test, and categorical variables were analysed using the Chi-square test or Fisher’s exact test.
Table 3. Multivariable logistic regression of same-day discharge.
* Statistically significant.
Discussion
Analysis of outcomes in Fontan patients undergoing non-cardiac procedures has been limited to adult patients Reference Rabbitts, Groenewald and Mauermann8,Reference Egbe, Khan and Ammash9 or patients undergoing high-risk surgery such as spine surgery. Reference Uehara, Takahashi and Akazawa10,Reference Przybylski, Hedequist and Nasr11 We present an analysis of the largest series of children and young adults with a Fontan circulation undergoing a large variety of non-cardiac procedures. In our cohort, in 125 (36.3%) of anaesthetic, patients were successfully discharged the same day without readmission at 24 hours.
A lower number of chronic medical conditions was associated with same-day discharge (OR 0.91 per additional chronic condition, 95% CI 0.76–0.98, p = 0.022). The contribution of medical disease is clearly important. A recent study showed that patients with CHD and at least 1 chronic condition had a greater length of stay (21 days; IQR, 5–68) than those with CHD only (9 days; IQR, 3–46). End-organ dysfunction is common among patients with Fontan physiology, and this can play an important role in both drug metabolism and overall rate of recovery.
The use of inotropes was an independent predictor of need for admission, while the amount of crystalloid received was not associated in the multivariable model. While echocardiography can identify systolic dysfunction, we did not have enough information to assess the diastolic dysfunction which can be challenging in single-ventricle patients. Reference D’Udekem, Iyengar and Cochrane13 The use of inotropes is a decision made by the anaesthesia team in order to maintain systemic blood pressures. As such, it may be a surrogate for overall poor circulatory status. It is also likely that the need for inotropes would suggest to the treating clinicians that they are less likely to be successfully discharged home.
Patients who underwent major procedures were more likely to be admitted postoperatively, but this would be expected of patients without significant cardiac comorbidities as well. Interestingly, we did not find any association with the type of airway technique employed, the use of neuromuscular blockade, or the anaesthesia duration to be associated with hospital discharge. This supports the contention that the best technique for anaesthetizing a patient with Fontan circulation is unlikely to be a “one size fits all” strategy but rather a technique tailored to the patient’s current physiologic condition, and the procedure being performed.
Due to differences in patient demographics, it is difficult to compare our results to other published series. In a study from the Mayo Clinic, there were 39 anaesthetics that were delivered with a 31% complication rate and 1 postoperative death. Reference Rabbitts, Groenewald and Mauermann8 However, the Fontan patients in this study were all adults of which 61.3% had an atriopulmonary connection, a form of cavopulmonary connection known to be associated with increased morbidity and mortality independent of non-cardiac surgery encounters. A larger follow-up study from the Mayo Clinic of 154 adult Fontan patients found a 12% intraoperative complication rate and a 5% postoperative complication rate including one death (0.2%). Reference Egbe, Khan and Ammash9 In these adult patients, however, only 13% underwent a general anaesthetic which is very different from our population of children and young adults with Fontan physiology. In a small study of 36 patients with hypoplastic left heart syndrome palliated to a Fontan undergoing non-cardiac surgery, adverse events consisted of respiratory instability in 3 (8%) and airway obstruction or laryngospasm in 2 (6%) patients. Of note, there were no cases of cardiovascular instability.
We do not have strict institutional guidelines for same-day discharge, hospital admission, or ICU admission but rather rely on clinical judgement of the primary cardiologist, the consulting cardiac anaesthesiologist, and the primary anaesthesiology team. Patients with cardiovascular disease who undergo non-cardiac surgery or imaging at our institution are cared for by paediatric anaesthesiologists in the general operating rooms. While cardiac anaesthesiologists provide consultative service during the preoperative phase including medical optimisation, intraoperative management strategies, and postoperative disposition, it is the primary paediatric anaesthesia team that assumes responsibility for the care of these patients intraoperatively and in the recovery room.
Study limitations include all those inherent to a retrospective review. The decision process which determines patient disposition is difficult to ascertain in a retrospective study. Of note, patients at our institution are generally anticoagulated with aspirin alone and thus do not require admission for restarting anticoagulation. Specific information about anticoagulation was not collected. While this is a single-centre study which may limit generalizability, it is a large study that provides granular data unavailable in current registries and administrative databases. Because patients with Fontan procedures are not routinely catheterised immediately prior to non-cardiac surgery or imaging, no cardiac catheterisation data were collected or analysed.
Patients with Fontan physiology can successfully undergo same-day surgery under anaesthesia. The type of anaesthetic was not associated with discharge home. However, the presence of chronic medical conditions, undergoing a major surgical procedure, preoperative admission, and need of inotropic support during the procedure are associated with reduced odds of same-day discharge.
Acknowledgements
None.
Financial support
This research received no specific grant from any funding agency, commercial, or not-for-profit sectors.
Conflict of Interest
None.
Ethical standards
The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national guidelines on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008, and have been approved by the institutional committees (Boston Children’s Hospital Institutional Review Board).
