Perioperative management and considerations in pediatric patients with connective tissue disorders undergoing cardiac surgery

Marfan syndrome and Loeys‐Dietz syndrome are connective tissue disorders associated with cardiac and vascular disease. Patients often require surgical repair, but limited data exist to describe their perioperative management.


| INTRODUC TI ON
Connective tissue diseases are a relatively rare, heterogeneous group of disorders that include Marfan syndrome and Loeys-Dietz syndrome (LDS). Both syndromes convey a predisposition toward cardiac valvular and aortic aneurysmal disease, which may require initial surgical intervention during childhood or adolescence. 1,2 Some studies suggest that patients with LDS tend to have more aggressive aortic phenotypes than patients with Marfan syndrome that may necessitate surgery at an earlier age. 3 In particular, patients with LDS type I, as compared to type II, have a median age of first surgery in adolescence versus adulthood. 3 Other studies show that patients in both populations may have equally aggressive disease that requires surgery at a young age to prevent fatal complications, though death from disease complications tends to occur earlier and with greater frequency in the LDS population. [4][5][6] A review of the Pediatric Health Information System registry, which includes pediatric patients at 40 participating institutions, showed that 294 patients with Marfan syndrome under the age of 25 were hospitalized over an 11-year period. 6 Of these 294 patients, 213 (72%) required at least one aortic root and/or valve surgery, 51 (17%) required at least one mitral valve surgery, and 43 (15%) underwent operations on both valves during a single hospitalization. 6 Some evidence suggests that patients with LDS have a similar risk of aneurysmal disease but that aortic dissections may occur at smaller diameters than they do in Marfan syndrome patients with comparably sized aortic vasculature and may be affected more distally than the aortic root. 3,7,8 In addition to cardiovascular disease, patients often have other prominent comorbidities, including scoliosis and chest wall deformities, high-arched palate, obstructive sleep apnea (OSA), and global low muscle tone. 1,3,[7][8][9] Many of these common, noncardiovascular comorbidities have the potential to impact airway management, hemodynamic stability, intra-operative blood loss, and extubation planning while patients are under anesthesia care.
Much of the perioperative literature characterizing Marfan syndrome and LDS is surgical in nature, primarily describing novel surgical techniques and related outcomes. 6,[9][10][11][12] However, the body of literature describing patient care under general anesthesia is notably limited 1 and largely restricted to orthopedic and obstetric procedures in adult patients. [13][14][15] A few small case reports discuss the specifics of the anesthetic management of patients with Marfan syndrome and LDS undergoing cardiothoracic surgery. 16,17 Patients with LDS are even less well-studied, given that the genotype and phenotype were much more recently described than those for Marfan syndrome. 8,17,18 Our institution is a pediatric quaternary care center with a large referral population of pediatric and young adult patients with Marfan syndrome and LDS undergoing cardiac surgical repair of their valvular and aortic disease. Using an institutional database of perioperative patient data, we sought to describe this cohort of patients, which is larger than any similar cohort previously described in the pediatric cardiac anesthesia literature. For this study, we narrowed our focus to identify the preoperative characteristics of these patients that are often associated with increased risk under anesthesia, such as known difficult airway, younger age, or prior cardiac surgery. We also described the salient features of intra-operative management, such as blood product transfusion, cardiopulmonary bypass times, and rates of immediate postsurgical extubation.
Additionally, although patients with Marfan syndrome and LDS are commonly grouped together under the umbrella of connective tissue diseases, we aimed to identify the patient characteristics and comorbidities that distinguish these individual disease entities, which might enable pediatric cardiac anesthesia providers to better understand their patients with distinct connective tissue diseases.

What is already known
• Marfan and Loeys-Dietz are rare disorders that often require complex cardiac surgery during childhood.

What this article adds
• This paper discusses the similarities and differences between patients with Marfan and Loeys-Dietz undergoing cardiac surgery.
• The basic demographics and potential anesthetic considerations for these patients are also evaluated, including the frequency of difficult airway (20% in Loeys-Dietz), patient age, and type of surgery.

Clinical implications
Marfan syndrome and Loeys-Dietz syndrome (LDS) are connective tissue disorders with cardiac and vascular disease that often requires cardiovascular surgical interventions. Surgical techniques, complications, and outcomes of these interventions have been studied extensively. Little is known, however, about the comprehensive anesthetic care of these patients. This study examined a cohort of patients with Marfan syndrome and Loeys-Dietz undergoing cardiac surgery at a single center. Using this large dataset, we identified several salient features that these groups have in common and that set them apart as distinct disease processes, all of which have the potential to affect their anesthetic management. This study shows that patients with Loeys-Dietz have a 21% chance of having a difficult airway at the time of cardiac surgery. Further, we show that patients with Loeys-Dietz tend to be younger at the time of first cardiac surgery, with a higher proportion having had more than one prior cardiac surgery.

| Study design
We conducted a retrospective observational cohort study of all patients between 0 and 21 years of age with a diagnosis of Marfan syndrome or LDS who underwent cardiovascular surgery at our institution between 2010 and 2017. If a patient underwent more than one operation during this period, each surgery was counted as a separate patient encounter. This study was approved by the Johns Hopkins Institutional Review Board with a waiver of consent. The Figure 1 was self-made by a member of our team.

| Data collection
We collected electronic medical record data from preoperative outpatient appointments, preoperative ward, or intensive care unit (ICU) for those patients admitted before surgery, the operating room, and the immediate postoperative ICU admission. The intra-and postoperative records were reviewed for duration of cardiopulmonary bypass and aortic cross-clamping, type and quantity of blood products and factors administered, use and maximum dosage of vasoactive medications, whether the patient was extubated in the operating room, peak intra-operative lactate levels, and first postoperative hemoglobin in the ICU.

| Statistical analyses
We used descriptive statistics to analyze the demographic and perioperative characteristics of the study population. We report continuous variables as medians with interquartile ranges (IQR) and categorical variables as counts with percentages. Bivariate group comparisons were made with t tests and Wilcoxon rank sum tests for continuous variables and chi-square tests for categorical variables. .2826
The total valve score is a variable that we created to estimate the valvular abnormalities in each population. Mild, moderate, and severe stenosis or regurgitation were given scores of 1, 2, and 3 respectively. The total valve score for patients with Marfan syndrome was 3 and for LDS was 1 (p = .029).
The majority of surgical procedures included valve-sparing aortic root replacements (79% for Marfan and 63% for LDS). Most other surgical procedures included repair or replacement of the mitral and/ or aortic valves. Surgical factors, including duration of cardiopulmonary bypass and aortic cross-clamping, were similar between groups.
"Difficult airway" management was more common in the LDS group.
Patients in the two groups received vasoactive medications, including epinephrine, vasopressin, and milrinone, with similar frequency.
Blood product transfusion was rare overall, though patients with

| DISCUSS ION
In this retrospective observational study, we recorded data on a relatively large cohort of 71 children, adolescents, and young adults with Marfan syndrome or LDS who underwent a total of 77 cardiovascular surgical procedures.
To our knowledge, this study is the first of its kind to describe the intra-operative course of this size cohort of patients with con- bleeding, arrhythmias, respiratory failure, and death). 6,[9][10][11][12]20 Many of the surgical studies do have the benefit of decades of experience with robust long-term follow-up. However, only a few studies conducted at large-volume children's hospitals or derived from pediatric healthcare databases are specific to children and young adults. 6,9,12,20 Of the surgical studies focused primarily on children, even fewer dedicate equal weight to patients with LDS as compared to Marfan syndrome. 11,12 At present, most peer-reviewed publications concerning the an- Cardiac and aortic disease is of great concern for all patients with connective tissue disorders who must undergo general anesthesia, but it is of particular relevance for cardiovascular surgery. 1,7,23 Most publications include specific recommendations for special populations, including pregnant women, but not for children. 1,7 Recommendations for care of pediatric patients with Marfan and LDS under anesthesia are supported mostly by anecdotal evidence from single-patient encounters or small case series describing anesthetic methods and observed complications. [16][17][18]22,23  vital capacity below 80% of predicted for age in 9 of the 13 patients.
Six had normal cardiovascular anatomy and physiology preoperatively. As in our study, none of the patients proved difficult to intubate. Anesthetic outcomes were mostly favorable intra-operatively, with a few exceptions, including one patient who suffered a cardiac arrest after induction of general anesthesia and one who developed a malignant arrhythmia leading to abortion of the surgical procedure. The benefit of reporting these data, including from smaller patient samples, is that it provides a glimpse into a wide range of patient phenotypes and anesthetic practices, which can be beneficial in demonstrating that variation in anesthetic technique can be safe and effective.
Additionally, single-patient reports may highlight rare presentations that pose unique challenges, as in a case of Marfan syndrome comorbid with Turner syndrome. 23 Difficult intubation was also largely absent in small case reports, [16][17][18]23 and laryngeal mask airway was successfully used for a minor procedure, 23  Our study has limitations, including that it is retrospective and relies on data from an electronic medical record. Therefore, it is subject to omissions and human error. Our sample size is large compared to anything previously published on this patient population, though it is still relatively modest. We also cannot account for the impact of any changes in surgical technique or materials. We recognize that the patient's immediate postoperative care in the ICU can be considered an extension of their operating room care and may provide more insight into postanesthetic outcomes. Data from this time period were collected for future analysis, but exist outside the scope of this present study.
Cardiovascular surgical techniques to manage the valvular and aortic comorbidities of Marfan syndrome and LDS continue to improve based on decades of follow-up data, permitting the ongoing care of patients with more severe disease. By collecting data and reporting on anesthesiologists' experience caring for patients with these diseases, anesthesia care will be able to evolve in tandem as patients are diagnosed earlier and require aggressive management of cardiovascular comorbidities at younger ages. Analyzing large databases of pediatric patients who undergo general anesthesia for cardiovascular procedures remains critically important for improving the safety of perioperative care.

DATA AVA I L A B I L I T Y S TAT E M E N T
Due to medical and ethical concerns, the data that support the findings of this study are available from the corresponding author, DG, upon reasonable request.