PE is characterized by a concave depression of the anterior chest wall of variable severity and can be symmetric ( Fig. 19.1A ) or asymmetric ( Fig. 19.1B ). Deformities can be characterized as cup shaped, diffuse, or eccentric (a combination of cup shaped and diffuse) ( Fig. 19.2 ). The depth and extent of the depression determines the degree of cardiac and pulmonary compression, which generally coincides with the degree of physiologic dysfunction.

(A) Symmetric pectus excavatum defect. (B) Asymmetric pectus excavatum defect.

(A) The typical focal or “cup-shaped” pectus excavatum deformity. (B) A wider, more diffuse or “saucer-shaped” deformity.

This chest wall deformity may be noted soon after birth ( Fig. 19.3 ), but infants and small children are generally asymptomatic, and caregivers should be reassured that immediate surgery to correct the chest wall deformity is not necessary. However, primary providers should be encouraged to refer patients early to establish care with a team with expertise in chest wall deformities in order to formulate a long-range care plan, as in most cases, the deformity does not spontaneously resolve. In most cases, patients are referred to a surgeon after a pubertal growth spurt precipitates significant progression of the defect, and options for treatment are sought.

Pectus excavatum may be seen in infants. This 11-month-old child was evaluated for pectus excavatum. Observation was recommended.

In 1594, Bauhinus described a patient with severe PE, dyspnea, and paroxysmal cough, which he attributed to pulmonary compression. Numerous other case reports appeared in the 19th century. The familial predisposition was first noted by Coulson in 1820, who cited a family of three brothers with PE. Treatment at that time consisted of “fresh air, breathing exercises, aerobic activities, and lateral pressure.” The earliest known (unsuccessful) attempt at surgical correction was described by Meyer in 1911 who removed the second and third costal cartilages on the right side. In 1913, the thoracic surgical pioneer, Ferdinand Sauerbruch, described an operation whereby he excised a section of the anterior chest wall, including the left fifth to ninth costal cartilages as well as a segment of the adjacent sternum in a patient with severely limiting dyspnea. After recovery, the patient successfully returned to work and reported his dyspnea had resolved. In the 1920s, Sauerbruch described the first PE repair that employed bilateral costal cartilage resection with sternal osteotomy later popularized by Ravitch. He also used external traction to hold the sternum in its corrected position for 6 weeks after surgery. In 1939, Ochsner and DeBakey published their experience with this approach. In 1939, Lincoln Brown published a review of the literature with particular reference to theories about the etiology of PE and believed that short diaphragmatic ligaments and the pull of the diaphragm on the posterior sternum were causative factors. Ravitch initially subscribed to this theory, and as a result, he advocated for an even more radical mobilization of the sternum, with transection of all sternal attachments, including the intercostal bundles, rectus muscles, diaphragmatic attachments, and excision of the xiphisternum. In 1949, he published his experience with eight patients in which he used this radically extended modification of Sauerbruch’s technique, but without external traction. To combat the potential for recurrence without external traction, Wallgren and Sulamaa introduced the concept of internal support in 1956 by using a slightly curved stainless-steel bar that was inserted through the caudal end of the sternum from side to side to bridge the gap between the sternum and ribs. In 1961, Adkins and Blades modified internal fixation further by passing a straight stainless-steel bar behind, rather than through, the sternum. As early as 1958, Welch and Gross advocated a less radical approach than that of Ravitch. Welch reported excellent results in 75 patients without dividing all the muscle attachments. However, he still advocated performing the procedure in young patients. Conversely, Pena and others at the time were very concerned about the idea of resecting the cartilages from very young children and demonstrated that asphyxiating chondrodystrophy developed in baby rabbits after cartilage resection during their growth phase. This sentiment has been confirmed by more modern work by Kanat and colleagues. Later, Haller reported the risk of acquired asphyxiating chondrodystrophy following operative correction in young children. As a result, many surgeons stopped performing PE repair in young children and waited until the child reached puberty. They also reduced the amount of cartilage resected and spoke about a “modified Ravitch procedure,” which was really the original operation described by Sauerbruch.

In 1998, Donald Nuss and colleagues described a departure from the decades-long practice of “open” PE surgical repair. In this approach, a convex steel bar was placed under the sternum using two small lateral chest wall incisions to allow for chest wall (cartilage) remodeling. The initial description included 42 patients under the age of 15 who underwent the new procedure, and the results were quite promising—with 22/30 children exhibiting excellent results after bar removal 2 years following original insertion. Now, some 25 years after its initial description and countless modifications (most notably using thoracoscopy), the minimally invasive repair of PE (MIRPE) has become the mainstay of surgical correction for most patients.

The incidence of PE ranges from 6.3 to 12 cases per 1000 persons. Even if PE occurs sporadically, a genetic predisposition seems likely, since a positive family history can be elicited in up to 43% of PE cases. ^, Within familial cases, inheritance is variable, being autosomal dominant, autosomal recessive, X-linked, and multifactorial in different families. Males are affected four times more commonly than females, but the latter have an increased risk of associated scoliosis.

The underlying etiology of PE is not definitively known. Theories cited previously by Brown and Ravitch are no longer widely supported. Investigators now suspect that alterations (increased flexibility, abnormal collagen, altered growth rates, or other abnormal components or structure) of the costal cartilages play a primary role, although no single explanation has been definitively proven. There is now a wealth of evidence to suggest that cartilaginous overgrowth is unlikely to be causative. In fact, several authors have noted that cartilaginous lengths are relatively shorter in patients with PE—the so-called undergrowth hypothesis. Alternatively, the fundamental cause of PE may be an abnormality in the composition of the cartilage, leading to greater flexibility. Nearly all of the published histologic studies to date have uncovered disturbances in the composition and structure of the extracellular matrix of costal cartilages (type II collagen) in patients with PE, and this theory becomes even more intriguing when considering the association between PE and connective tissue disorders such as Marfan or Ehler-Danlos syndromes. The association with a connective tissue disorder is higher than in the normal population. Most patients have a tall, thin, asthenic build, and whereas up to 15%–20% of patients may have clinical features suggestive of Marfan syndrome, a smaller percentage (only 2%–3%) have the diagnosis confirmed. Some 32 syndromes have been described with PE/PC as a frequent feature and 27 syndromes as an occasional feature. Mild scoliosis is present in upward of 25% of patients, and correction of the pectus often will improve mild scoliosis.

It is important to note that in both sporadic and syndromic cases, the abnormal chest configuration is typically not as prominent in early childhood and becomes more pronounced during pubertal growth. Thus, it seems plausible that a combination of abnormalities of the costal cartilages coupled with directional stress related to growth or other functional directional forces results in the progressive worsening of the pectus defect during the pubertal growth spurt.

PE may be noted in infancy and usually progresses slowly as the child grows. Most patients have a relatively mild PE deformity during childhood ( Fig. 19.3 ). Although the deformity may not always increase in severity, it is unlikely that it will spontaneously resolve. Because children have significant cardiac and pulmonary reserve and their chest wall is still very pliable, most young children are asymptomatic. However, with rapid pubertal growth, the deformity typically becomes more severe, and a mild condition may become severe in as little as 6–12 months. Eventually, many patients report difficulty keeping up with their peers when engaging in aerobic activity. The earliest symptoms are shortness of breath and lack of endurance with exercise. As the PE progresses, chest pain and palpitations with exercise can develop, giving rise to exercise intolerance and avoidance. A vicious cycle may develop as patients stop participating in aerobic activities altogether, and exercise capacity diminishes further. Patients with a rapid progression of their deformity seem to exhibit the most pronounced symptoms.

Many patients will avoid situations in which they have to expose their chest wall to others, which may limit participation in school and team activities. PE patients often have a poor body image, which has a major impact on their self-worth. Many PE patients have a typical geriatric or “pectus posture” that includes thoracic kyphosis, forward-sloping shoulders, and a protuberant abdomen ( Fig. 19.4 )—perhaps an unconscious attempt to hide the chest wall deformity from others. By withdrawing from participation in activities with their peers, they also may become depressed, which can affect their schoolwork and daily functioning.

This boy has the classic pectus posture with thoracic kyphosis, forward sloping shoulders, lumbar lordosis, and lower rib flaring (mild). There is often a protuberant abdomen or “potbelly” appearance as well.

Just as one would not consider leaving a child with a cleft lip untreated, one should not leave a child with a severe PE untreated. Both have a physiologic and psychological impact on the patient.

Although once controversial, it is now well established that severe PE can have a significant detrimental effect on cardiopulmonary function. There is a correlation between the severity of the defect and the degree of impairment, but individual subjects often do not perceive a limitation. Cardiac effects from PE include decreased cardiac output, mitral valve prolapse, and arrhythmias. Compression of the heart results in incomplete filling and decreased stroke volume, which results in decreased cardiac output. This was nicely illustrated by a study examining intraoperative echocardiography prior to and after sternal elevation in 168 adult patients undergoing MIRPE repair. Cardiac compression may also interfere with normal valve function. Mitral valve prolapse has been found in a disproportionate number of patients with PE (from 13% to 65%), and this may resolve with surgical correction. ^,

Pulmonary effects likely result from poor motion of the depressed part of the chest wall. Normal chest wall motion includes a “pump handle” movement of the sternum. The lower sternum moves up and out, like the handle of a mechanical water pump. Physical examination, and now motion capture analysis, show that this motion is almost absent in the depressed area of the PE chest. Instead, patients compensate by abdominal diaphragmatic breathing. Following correction, this motion is indistinguishable from the normal chest wall. Pulmonary function testing shows statistically and clinically meaningful diminution in static pulmonary function tests in uncorrected PE (forced vital capacity, forced expiratory volume in one second) even though the airway and pulmonary parenchyma are anatomically normal. It is interesting to note that although meta-analyses like that of Malek have not shown an improvement in pulmonary function after repair, the majority of the reports reviewed were on patients who underwent the Ravitch technique. Conversely, several modern reports have demonstrated improvement in cardiopulmonary function after the MIRPE operation. ^,

The general physiologic consequences of PE appear to be a reduction in the ability to perform aerobic exercise. Stress testing has shown an increase in oxygen consumption for patients with PE for a given exercise when compared with normal patients. Furthermore, patients undergoing stringent exercise testing have been noted to have a reduction in VO ₂ max (the maximal amount of oxygen the subject can take in via the lungs and transport to the skeletal muscles for aerobic metabolism), which is maintained into adulthood. These values improve with surgical correction, especially after MIRPE.

A complete history and physical examination is performed on all patients and includes documenting AP and lateral photographs both standing and supine. Special elements of the history include a family history of chest wall deformities, known family members with connective tissue disorders or a personal history thereof, a history of metallic allergies, and associated symptoms—especially as it relates to shortness of breath, exercise intolerance (relative to peers and previous stamina), and self-confidence and depression or psychosocial withdrawal. Apart from a complete examination, specific notation should be made to the characteristics of the defect (severity, type, sternal tilt, and symmetry), associated scoliosis, degree of costal flaring, and an assessment of marfanoid characteristics.

Infants and toddlers with PE defects are generally observed, as the majority are asymptomatic. Younger children with a mild to moderate deformity are treated with a posture and exercise program (physical therapy) in an attempt to halt the progression, and are followed at yearly or longer intervals ( Fig. 19.5 ). In our experience, younger patients may be candidates for vacuum bell therapy (see below). The severity of sternal depression has often been characterized based on the measurement of the “Haller index” (HI), which gives an objective measurement for comparing severity among patients. This is typically calculated using axial imaging: computed tomography (CT) or magnetic resonance imaging (MRI). The HI is calculated by dividing the transverse diameter of the chest wall by the anteroposterior diameter at the deepest point of the defect ( Fig. 19.6A–E ). An HI greater than 3.25 has generally been accepted as a severe deformity (normal being less than 2.5). Multiple authors have noted flaws in the traditional measurement schemes and have devised other systems to objectively measure anatomic severity including the pectus correction index (PCI), the Titanic index, and the optical 3D index. The major limitation of the HI is the transverse diameter of the chest: a patient with a wide chest may have a spuriously high HI, and conversely, a patient with a deep, narrow chest may have a lower HI than one may have predicted based on physical examination. The calculation of the PCI is described in Figure 19.6. The definition of PE was a PCI of greater than 10% of the chest depth lost at the inferior margin of the sternum. The initial study did not suggest that any specific PCI was an indication for repair, rather that a PCI of 10% accurately distinguishes normal chests from those with PE. Poston and colleagues tried to identify a PCI that would reach the threshold for recommendation for repair (based on an HI of >3.25) and suggested the threshold for repair was a PCI >28%. However, there is currently no agreed upon threshold of PCI to use as a standalone trigger to recommend repair, as the recommendation for such also depends on other factors. Practically speaking, most reimbursement agencies in the United States still rely on the HI calculation to determine “severity” as a first step in determining if they will pay for surgical correction. Thus, we and others remain dependent on the use of the HI in the preoperative evaluation of PE patients, although the PCI is gaining more acceptance.

Algorithm for evaluation and treatment of patients with pectus deformities.

PFT, Pulmonary function tests.

(A) CT scan showing cardiac compression and displacement, pulmonary compression, asymmetry of the chest, and sternal torsion. (B) CT scan showing severe pulmonary compression and atelectasis. (C) CT index is calculated by dividing the transverse diameter by the anteroposterior diameter. This compares the measurement of (D) the Haller index (HI) and (E) the correction index (CI). Although a PE deformity is clearly present in this patient with a deep chest, the HI is normal, while the CI is not.

The CI is calculated by drawing a horizontal line across the anterior spine. Two distances are then measured: the minimum distance between the posterior sternum and the anterior spine, as is utilized for the HI, and the maximum distance between the line placed on the anterior spine and the inner margin of the most anterior portion of the chest (on the same CT slice chosen for the minimum distance). The difference between the two lines is the amount of defect the patient has in their chest. To calculate the percentage of chest depth the patient is “missing,” take the difference between the measurements, divide by the maximum prominence of the chest (the longer measurement), and multiply by 100. The definition of pectus excavatum was a CI of greater than 10% of the chest depth lost at the inferior margin of the sternum.

From St Peter SD, Juang D et al., A novel measure for pectus excavatum: the correction index. J Pediatr Surg . 2011; 46(12):2270–2273. https://doi.org/10.1016/j.jpedsurg.2011.09.009 .

Patients with a severe PE deformity or those with documented progression may benefit from ongoing physical therapy for exercise and postural improvement, but surgical correction should be offered. Patients then undergo objective assessments to evaluate whether their condition is severe enough to warrant repair. Some have adopted the use of cardiac MRI as the preoperative initial imaging study to coalesce the goals of determining the HI as well as to interrogate the cardiac system. The anatomic assessment of the HI has been shown to be equivalent to CT, and reports should be standardized to determine the impact of PE on cardiac function. Standard pulmonary function and cardiopulmonary exercise testing is also generally undertaken as a preoperative assessment tool to determine suitability for candidacy for surgical repair. ^,

Determination of a severe PE and the need for repair include two or more of the following criteria: (1) a CT index greater than 3.25, or correction index greater than 20%; (2) pulmonary function studies indicative of restrictive airway disease; (3) the presence of mitral valve prolapse, cardiac displacement or compression, or conduction abnormalities on the echocardiogram or ECG; (4) documentation of progression of the deformity with associated physical symptoms other than isolated concerns of body image; (5) a failed Ravitch procedure; or (6) a failed MIRPE. With these criteria, only about 60% of patients are found to have a deformity severe enough to warrant correction. Indications for operation in PE vary between surgeons/institutions.

The age parameters for operative correction depend on the type of repair selected. Unlike the more invasive procedures (i.e., Ravitch procedure, sternal turnover), there is no interference with growth plates with the MIRPE. Therefore, it can be done at any age as evidenced by the fact that successful correction has been reported in patients as young as 1 year of age. Furthermore, MIRPE has gained rather widespread adoption in the adult repair of PE. ^, However, the concern with patients younger than 11 years is that if the procedure is performed at too young an age, many years of subsequent growth remain after which PE can recur. Although many surgeons believe that the optimal age for MIRPE repair is between 11 and 14 years, others have advocated for repair as early as 4 years of age. These advocates of early repair have demonstrated low recurrence rates (less than 1%) with a low rate of costal flaring and excellent patient satisfaction and equivalent complication rates. Regardless of how early one advocates for repair, prepubertal patients have soft and malleable chest walls, there is generally a quick recovery with a rapid return to normal activities, and results are excellent. After puberty, the flexibility of the chest wall decreases, sometimes requiring the insertion of two bars, which makes the operation and recovery more difficult. However, several centers have reported success with MIRPE into the fourth decade of life. ^,

Preoperative counseling should include a discussion of other surgical options, and the modified Ravitch repair may be a suitable option for those with mixed defects, those with significant sternal tilt, for older patients, or simply for those patients who would rather undergo the Ravitch procedure after a thorough discussion of the risks and benefits of both operative approaches.

Younger patients with a mild to moderate deformity are generally treated with a posture and exercise program in an attempt to halt the progression and are followed at yearly or longer intervals ( Fig. 19.5 ). One option that we have found to be suitable for patients with moderate defects or young patients (i.e., cooperative children prior to 11 years of age) is the vacuum bell device. Developed by Klobe, this offers a noninvasive method of treatment for patients who have mild to moderate disease or who opt not to have surgical correction ( Figs. 19.7 and 19.8 ). While vacuum bell therapy may not achieve the immediate and impressive improvement after a MIRPE or Ravitch procedure, it eliminates the inherent operative risks. Several studies have also demonstrated improvements in deformity depth and HI with vacuum bell use, making it both a safe and potentially effective alternative form of therapy in select patients. ^,

This boy is being managed with the vacuum bell, which relies on suction to bring the anterior sternum forward and correct the pectus excavatum. It is currently being used more internationally than in the United States.

Photograph Courtesy Dr. Carlos Segura.

The vacuum bell was used in this child. On the left (A) is the appearance of the chest wall prior to treatment with the vacuum bell. On the right (B) the pectus excavatum is improved following several months use with the vacuum bell. Note the suction marks caused by the vacuum bell on his anterior chest wall.

Photograph Courtesy Dr. Carlos Segura.

A variety of measures have been used for pain management in the postoperative period. Once quite popular, epidural analgesia has now fallen out of favor due to inconsistent results and the potential for rare, yet devastating, neurologic consequences. Many surgeons now utilize a “multimodal pain management plan” using a combination of intravenous narcotics, benzodiazepines , nonsteroidal antiinflammatory agents, and regional anesthetic techniques (vertebral blocks, paravertebral blocks, and paravertebral and intercostal infusion catheters) and nonpharmacological modalities such as meditation, distraction, music, and others.

Cryoablation has emerged as another intriguing pain management intervention ( Fig. 19.12 ). Cryoablation uses the application of extremely cold temperatures to temporarily disrupt peripheral nerve function to provide analgesia. The surrounding fibrous neural sheath critical for nerve regeneration remains intact. The cryoablation probe delivers short bursts of extremely cold temperatures where the rapid expansion of pressurized gas results in rapid cooling of the probe tip. The adjacent intercostal vessels are protected as they serve as a thermal sink and rapidly disperse the localized temperature change. Axonal regeneration occurs over weeks to months. One must make provisions for protection of surrounding tissues (skin and lung) as the cryoprobe is applied to thoracic interspaces 4–7 in both chest cavities using bilateral thoracoscopy, as described above. There is also a small potential for residual neuropathic pain and Horner syndrome. In two retrospective reviews of patients who underwent the MIRPE for PE, compared with a thoracic epidural, cryoablation was associated with lower opioid use and shorter length of stay and days to transition to oral pain medication but at the cost of longer operative time. ^, Although some studies have reported increased rates of bar migration/slippage in cryoablation patients—between 8% and 12% (perhaps due to increased postoperative movement/activity from lack of pain)—Lai and coauthors noted a bar displacement rate of less than 1%. ^, This group reported their experience with cryoablation in 350 consecutive patients and reported a 74% decrease in opioid use and an average length of stay of 2.7 days. This modality has also been used in patients undergoing the open repair of PE with similar benefits. ^, In another study of 110 patients who underwent cryoablation with MIRPE, 44% completed a discharge survey to ascertain pain control adequacy, with 55% of patients reporting tolerable residual pain at 2 weeks and 41% at 3 months, with 25% requiring intermittent pain medication at 3 months. Although cryoablation clearly is an excellent pain control modality, patients likely underreport functional symptoms and experience more frequent discomfort and alteration of daily living activities than may be reflected in reports of cryoablation “success.” Whatever type of pain management is selected, utilizing strategies to decrease narcotic usage has been a major accomplishment over the last decade.

The use of cryoablation significantly reduces/eliminates pain in the postoperative period. On the left (A) the cryoablation probe is being applied to the intercostal nerve bundle below the third rib on the right side. It takes 2 minutes to cryoablate each intercostal bundle. Note the ice forming along the cryoablation probe. On the right (B) the probe has been inserted into the right chest through the lateral chest wall incision. Note the cannula and telescope that have been introduced into the right thoracic cavity through the same lateral chest wall incision. It is important to retract the inferior aspect of the incision so as not to freeze the skin as well.

MIRPE received rapid acceptance by the surgical community because the technique requires neither rib resection nor sternal osteotomy. ^, Blood loss is minimal, the operating time is short, and patient satisfaction is generally excellent. Although the initial report from Nuss and colleagues in 1998 presented a 10-year experience, the numbers were limited, and the long-term results were affected by an early learning curve and using a pectus bar that was too malleable. Moreover, in some of these patients, the bar was removed too early. Since that time, the Nuss group and many others throughout the world have reported their experience with MIRPE. Numerous important modifications have been made, both to the operative technique (e.g., routine use of thoracoscopy) and instruments, to minimize the risks of the operation and facilitate insertion and stabilization of the support bar. These modifications have markedly reduced the risks and complications. An impressive amount of experience with the procedure has brought about generally excellent results. ^{, ,} The cardiopulmonary improvements after MIRPE have been addressed previously, but MIRPE has shown the best cardiopulmonary improvement compared with other surgical techniques. There is evidence of an improvement in pulmonary function (normalized FEV1) following MIRPE but not after a Ravitch repair. Improvements in both physical and psychosocial function have been reported after surgical repair, which are generally durable even after bar removal.

Patients are evaluated at regular intervals following the operation. We generally obtain a chest radiograph 3 months after surgery to document adequate bar position, but routine chest radiography thereafter is discouraged unless there is a clinical indication. We recommend a pectus bar “dwell” for 3 years if possible. We evaluate patients on a semiannual basis and monitor their growth and activity levels and encourage them to perform their physical therapy exercises and participate in aerobic sports if desired. Patients between the ages of 6 and 10 years often do not grow rapidly. Therefore, they tolerate a bar dwell of 3 or even 4 years. Conversely, teenagers may experience a massive growth spurt, completely outgrowing the bar, and the bar may require removal sooner than 3 years—typically when it begins to cause discomfort due to chest wall compression. Exercise programs are an important adjunct to success. Many children and adults lead sedentary lifestyles and never perform aerobic activities. Therefore, their lungs never expand beyond the resting tidal volume (approximately 10% of total lung capacity). Deep breathing with breath holding for 10 to 15 seconds and aerobic activities, such as running (e.g., soccer, basketball) and swimming are strongly encouraged. Most bar removals are uneventful and managed as an outpatient procedure. However, complications can develop including significant hemorrhage, wound infection, hematoma ( Fig. 19.14 ), and even cardiac arrest and death. When performing bar removal, provisions should be made for life-threatening hemorrhage, and instrumentation and personnel should be readily available in the event of a catastrophic event.

This patient has undergone removal of the bar, but a hematoma developed around the right lateral chest wall incision after bar removal. The hematoma resolved by wrapping the chest wall with an elastic bandage.

Photo Courtesy Dr. Michele Ugazzi.

Long-term assessment has allowed classification of the results into excellent, good, fair, or failed categories. An excellent repair indicates that the patient experienced complete repair of the pectus deformity and resolution of the associated symptoms. A good repair is distinguished by a markedly improved but not completely normal chest wall appearance along with resolution of the associated symptoms. A fair result indicates a mild residual PE without complete resolution of symptoms. A failed repair is defined as a recurrence of the pectus deformity and associated symptoms, or the need for additional surgery (or both) after final removal of the bar. Results have been excellent or good in greater than 95% of patients.

PC is the other main category of congenital chest wall abnormality. The name derives from the Latin term for the keel of a ship (“ carina ”). Although excavatum and carinatum deformities are described separately, in reality, there is often overlap. Defects that are mostly sternal depressions may have areas of protrusion and vice versa. The most common form of carinatum is chondrogladiolar ( Fig. 19.15 ). Gladiolus means shield and refers to the lower portion of the sternum. This type accounts for 90%–95% of cases. The remainder are mostly mixed defects, with about 1% being primarily chondromanubrial protrusions, known variously as Currarino-Silverman syndrome, pectus arcuatum, or “horns of steer” ( Fig. 19.16A–C ). This defect is usually not amenable to bracing, and the surgical approach is more complex than for the common forms of PE or PC.

These images depict a pectus carinatum. (A) A patient with a chondrogladiolar (90% of cases) pectus carinatum. (B) A lateral chest radiograph.

(A–C) Chondromanubrial protrusions, known variously as Currarino-Silverman syndrome, pectus arcuatum, or “horns of steer.” This defect is usually not amenable to bracing, and the surgical approach is more complex than for the common forms of PE or PC. (A) Lateral view of the chest. (B) Anterior view of the chest demonstrating upper protrusion and the appearance of an excavatum (depression) just below the abnormal protrusion. (C) Lateral chest radiograph of the deformity.

The traditional surgical repair of PC is a Ravitch procedure, first described in 1952. Orthotics were used as an alternative, with the initial report in 1979 and a later series (37 patients) by Haje and Bowen in 1992. A larger series using a dynamic compression system of bracing was reported by Martinez-Ferro from Argentina. Bracing is currently the most common treatment for most children, and the availability of an alternative to surgical correction has seen an increase in the numbers of PC patients presenting to the pediatric surgeon.

The incidence of PC ranges from around 1 in 1500 to 1 in 5000 live births, but these numbers can vary widely depending on the population studied and the methods used to diagnose the condition. It was previously thought to be much less common than excavatum (one-fifth as frequent), but with the advent of nonoperative treatments, the relative incidence (at least that seen clinically) has increased. PE is still about two to three times more common than protrusion deformities in the United States, but substantial geographic variation exists. PC shares a male to female ratio with excavatum (about 4:1). A family history is frequently present, seen in one-quarter to one-third of patients.

Common associations include scoliosis (usually thoracic, and between T4 and T9) in about one-fifth of children; 14% of 131 children with pectus carinatum required treatment for their spinal curvature in an older series of patients with scoliosis and PE or PC. Clinical examinations for scoliosis vary in sensitivity, particularly with less severe curvature. A high index of suspicion and the use of spine radiographs and orthopedic consultation (when indicated) is necessary.

Other known associations include connective tissue disorders (Marfan or Ehler-Danlos), Turner syndrome (45×), Poland syndrome (discussed later), and Noonan syndrome. There are multiple other less common/rare monogenic syndromes associated with carinatum and/or excavatum, including neurofibromatosis, Morquio disease, PHACE (posterior fossa brain malformations, facial hemangioma, arterial abnormalities, cardiac defects, and eye problems), and osteogenesis imperfecta. ^,

Marfan syndrome is an autosomal dominant connective disorder due to mutations in the Fibrillin-1 gene (FBN1) on the long arm of chromosome 15. Most patients with Marfan syndrome have scoliosis. Excavatum defects are more frequent than carinatum abnormalities in children with Marfan syndrome, but both are common. They may be asymmetric. Conversely, in a large series of children with PC, the percentage with Marfan (or other syndromes) is very small (e.g., 1.2% with Marfan syndrome in a series of 430 children from 2021). ^,

Turner syndrome is common (1 in 2500), and most have sternal abnormalities (PE/PC), albeit often not requiring correction. These appear to be independent of the cardiac defects (bicuspid aortic valve and aortic coarctation).

Children with Noonan syndrome have characteristic facies, short stature, congenital heart defects (often pulmonic valvular stenosis and cardiomyopathy), and variable developmental delays. Other findings can include superior PC and/or inferior excavatum, as well as a variety of other abnormalities. Some chest wall abnormality is seen in most Noonan patients.

The frequent presence of a family history of chest wall deformity as well as the markedly increased incidence in children with known genetic abnormalities (e.g., Marfan syndrome) suggests an inherited component, although the precise etiology remains unknown. Abnormal costal cartilage, rather than an intrinsic sternal abnormality, is felt to be responsible for the visible deformity. Rarely, PC can occur as a postoperative complication following median sternotomy or prior PE repair. ^,

In contrast to PE, carinatum defects are usually identified later, often with pubertal growth spurts. The cosmetic appearance with its attendant negative effects on social interaction and perceived body image are the most common reasons for surgical consultation. Diminished disease-specific and health-related quality of life is well documented in these children, along with the voluntary restrictions in their social activities. Chest (or less often, back) pain is not infrequent, usually unrelated to activity. An increased incidence of asthma is reported in association with PC, and other respiratory symptoms (dyspnea, shortness of breath, and fatigue with exertion) are common. However, as many as half of the patients are asymptomatic.

The diagnosis is clinical. A thorough history should be obtained, with documentation of all symptoms including the psychosocial effects (quitting sports, avoidance of swimming in the summer, withdrawal from activities). A family history of chest wall deformity and the affected relative (s) should be obtained, as well as any history of one or more of the associated syndromes. If a surgical procedure is contemplated, a history of metal allergy or atopy should be sought.

The physical exam should document the deformity (preferably with photographs in the electronic medical record). Lens abnormalities as well as any abnormal musculoskeletal findings are recorded, along with the general body habitus. Mitral valve prolapse may be associated, and auscultation of the chest and heart sounds is important. Scoliosis is a known association and shoulder or pelvic girdle tilt, and spinal curvature should be evaluated. An Adam’s forward bend test is helpful. An anterior-posterior and lateral chest x-ray are usually obtained. Pulmonary function studies are often obtained and may be particularly helpful in a patient with a mild chest deformity but respiratory symptoms out of proportion to the defect. An echocardiogram is routinely obtained by many but is clearly indicated for suspicious symptoms, a murmur, or associated abnormalities (Marfan or Noonan syndrome). In a study of 155 carinatum patients who underwent echocardiography, five patients (3.2%) had mitral valve prolapse, and 11 (7.1%) had aortic root dilation. Patient-reported symptoms were not significantly associated with abnormal echo findings. This suggests routine screening is a reasonable approach.

There was no effective treatment of PC until the 1950s–1960s, when Ravitch and others described surgical correction in case reports. ^{, ,} In the 1990s, nonoperative management via bracing gradually began to be used, and it is currently the initial treatment of choice for most children with PC. The success of the Nuss operation for excavatum may have helped to demonstrate that the chest wall was flexible enough to be reshaped, thus supporting nonoperative approaches to protrusion defects.

Bracing/Compression

Bracing can be static, for example with custom-made orthotic compression devices as initially described, or dynamic (adjustable during treatment based on compression pressures and patient response). Dynamic compression bracing (DCB), introduced by Martinez-Ferro in 2008, has become the standard.

Although most children with PC are candidates for bracing, there are exceptions. Chondromanubrial (Currarino-Silverman) defects almost always require surgical correction (discussed below). Mixed or asymmetric defects are often either not amenable to bracing or would be unlikely to have a satisfactory cosmetic result. The appropriate age for bracing varies in the literature, but there is a direct relationship between the PIC (pressure of initial correction, or force needed to flatten the carinatum defect) and age—older children and adults unsurprisingly have a stiffer chest wall. Bracing is therefore easier at younger age, but two factors result in the early teens being the preferred age for brace correction: (1) most defects are not noted or not clinically significant until rapid growth during puberty, and (2) the risk of recurrence is increased if bracing is started too early. ^,

The details of bracing vary significantly between institutions. ^, At the initial visit, assessment is done as noted above. The chest is measured, and baseline photographs are obtained. Newer handheld scanning devices allow quicker, more accurate measurement and a better fit of the brace ( Fig. 19.17 ). PIC is measured by having the patient stand upright with their back against a firm surface, such as a door. Several braces are available—we use a DCB from Pampamed (FMF, Buenos Aires, Argentina). ^, The DCB is ordered, and an initial fitting is scheduled ( Fig. 19.18 ). The brace is adjusted to use enough force to compress the chest while avoiding excessive pressures resulting in discomfort (and noncompliance) or skin/soft tissue injury. This pressure of treatment is usually set to approximately 2.5 PSI. The patient is advised to wear the brace all day and night as much as possible, with removal for sports, bathing, or showering. ^, Return visits to clinic are scheduled every 2 months, with repeated pressure measurement, brace adjustment, and photographs. Duration of treatment varies between patients and institutions, but poorer compliance, higher PIC, and increased age correspond to a longer duration of bracing needed for correction. ^{, ,} Discomfort with the brace is the most common reason for noncompliance, and noncompliance is by far the most common source of bracing failure.

Images from handheld three-dimensional scanning to determine chest contour and measurements for external bracing.

A patient being treated with dynamic compression bracing for a prominent pectus carinatum, (A) front view and (B) lateral view. The advantage of this brace is that it allows a preset pressure to be used to compress the carinatum.

Once full correction is obtained, a retainer mode is entered with gradual reductions in the DCB wear time and less frequent follow-up. Some patients (up to 40% of a large series in the Kansas City experience) failed to maintain improvement during this retainer mode (correction pressure of >1.0 PSI after previous correction). Most of these “recurrences” were mild and responded to a return to the 23-hour bracing schedule. The overall risk of recurrence in those who complete a bracing program has varied (depending on the definition and other factors), but it ranges from 2% to 25%. ^{, , , , ,} A recent systematic review cited a recurrence rate of 2.5% in nearly 1200 patients. Patient satisfaction is usually over 90%.

The current literature on bracing for PC consists of various retrospective series, with variations in treatment protocols, outcomes, and follow-up. A consistent issue is that of drop out, or failure of the patients to complete the bracing program. This occurs in 15%–40% of patients in most series. Common reasons include social discomfort, perceived failure of treatment, and pain or discomfort. ^{, ,} Studies have suggested that a lower initial PIC and a more rapid early rate of improvement, measured by faster decreases in the pressure of correction, correlate with a lower drop out/noncompliance rate. ^{, ,} More rapid improvement may provide positive reinforcement to continue bracing.

Skin and soft tissue complications are relatively minor and infrequent (8% in a systematic review). These include hematoma, skin abrasion or ulceration, and blistering. Overcorrection is rare.

Other bracing complications in a review of nearly 1200 patients from eight series included mild chest discomfort or tightness (12%), back pain (3.8%), chest pain (2.3%), tachycardia (0.8%), vasovagal episodes during brace fitting (1.0%), and paresthesia of digits on the hands (2.7%). Mechanical problems related to the brace occurred in 5.6% of patients.

Concern about decreased pulmonary function (after surgical repair of carinatum) prompted pre- and posttreatment pulmonary function studies and echocardiograms in a series of 61 patients, with findings of no effect on pulmonary function tests after bracing, and improved compliance.

Surgical experience with these defects is limited due to their relative rarity. An open surgical approach is almost always necessary. A vertical incision is used, with subperichondrial resection of the costal cartilages that attach to the sternum at and below the level of the fusion (including the second rib). A substernal plane is created beginning at the subxiphoid region. A wedge osteotomy is then performed that allows for flattening of the sternum. Additional osteotomies may be needed to flatten the sternum. Depending upon the length of the sternum on each side of the wedge osteotomy and the presence of additional osteotomies lower down, sternal fixations, either ladder or straight plates, are screwed into the sternum, bridging the fracture to provide stability to the chest wall immediately while the sternum is healing. It ensures that straight alignment of the sternum is preserved during healing while diminishing the risk of recurrence, asymmetry, or impaired healing. The hardware is typically completely covered when the pectoralis muscles are reapproximated in the midline.

Overall, outcomes from bracing and operation for PC are excellent. The availability of a noninvasive treatment (DCB) has resulted in increasing numbers of children and families seeking treatment. A reasonable approach is to offer bracing as first-line therapy whenever possible, with surgery reserved for unfavorable anatomy, bracing failure, or those who prefer surgical options to bracing. Long-term results of both are excellent, with a low incidence of complications and a high rate of patient satisfaction and an improved self-image. ^{, , , , ,}

Alfred Poland described absence of the sternocostal pectoralis major and pectoralis minor, hypoplasia of the serratus, and missing/abnormal digits in the left hand of a 27-year-old convict in 1841. However, there were documented cases prior to this. Somewhat controversially, the term “Poland syndrome” has been in use since 1962 when Clarkson reviewed an earlier report, by Brown and McDowell, that cited Poland’s earlier description. Critics of the current eponym point out that Poland was not the first to describe the anomalies and that he did not realize the two deformities (chest wall muscular deficiency and extremity lesions) even constituted a syndrome. ^,

Poland syndrome (PS) is a rare condition, affecting about 1 in 30,000 live births. It is almost always sporadic, although there are rare familial occurrences. It is more common in boys than girls—in a 2018 review of 245 patients, 155 (66.3%) were males and 90 (33.7%) females. It is more likely to affect the right side of the body (approximately twice as often, in a recent series of 113 patients).

PS is partial or complete absence of the pectoralis major. The sternocostal head of the pectoralis major muscle is uniformly absent, with the entire muscle missing or hypoplastic in most ( Fig. 19.19 ). ^, Other muscle deficiencies include (in descending order of frequency) absent or hypoplastic pectoralis minor (95%), serratus anterior, latissimus dorsi trapezius, rhomboid, and rectus abdominus.

Poland syndrome with right-sided absence of the pectoral’s major and minor, abnormal breast and areolar development, and rib hypoplasia. The contralateral side is completely normal.

Approximately 15% have rib aplasia or hypoplasia and, less commonly, involvment of the clavicle and sternum. Most patients have abnormalities in at least two ribs, usually the third and fourth. Lung herniation is seen in approximately 5%–10% of patients.

Some authors define “minimal” forms of PS as having only muscle or only rib abnormalities, but most require pectoral abnormalities plus at least one additional component to qualify for the diagnosis. ^,

In large series, approximately half the patients have an abnormality of the upper extremities. ^, Various classifications for the specific hand anomalies exist. One of the more common is the Catena-Senes classification (based on 175 patients), containing eight types. The most common abnormality is partial or complete symbrachydactyly (absence of middle phalanges), but syndactyly, brachydactyly, and extrodactyly (absent digits) are seen as well as absence or hypoplasia of carpal bones, metacarpals, or tendons; shortening of the humerus, radius or ulna; and even nail agenesis.

Sprengel deformity (congenital scapular elevation) is also seen in 5%–15% of PS patients and is usually associated with hypoplasia or absence of the serratus anterior muscle. ^{, ,} Scoliosis, vertebral abnormalities, and pectus deformities can also be seen.

Breast abnormalities include absence or underdevelopment of the breast or areola (amastia and/or athelia) ^{, , ,} Some degree of breast abnormality is seen in most females with PS. ^, Contralateral gynecomastia has been reported in males. ^,

Dextrocardia has been associated with left-sided PS in 12% of cases in a series of 122 patients, and it was more common when there was hypoplasia or aplasia of two or more ribs.

Möbius syndrome is characterized by congenital nonprogressive facial weakness (VII ^th cranial nerve palsy) and limited abduction of the eye (VI ^th cranial nerve palsy). The association with PS is well known, but it is rare—only 19 case reports since 1981 —and the underlying mechanism is unknown. About 15% of patients with Möbius syndrome have features of PS.