Pectus excavatum is the most common chest wall anomaly. It affects one in every 400 newborns in the United States,1 and is characterized by depression of the anterior chest wall and sternum. Impaired cardiac and lung function has been widely described in patients with severe deformities,2 and the chest deformity causes body image issues in affected individuals. Although more than 40% of our patients report a family history of chest wall deformities,3 the etiology of the disorder remains unknown.

With increasing clinical experience, we were able to present a classification of the dysmorphology of pectus excavatum.4 We have conducted collaborative genetic studies on this disorder, which showed that inheritance occurs via different pathways in different families: autosomal dominant, autosomal recessive, X-linked and complex hereditary patterns have been observed.5

Genetic analysis attempts to establish a correlation between genotype and phenotype.6 Previously, both we and other authors7 have described several variants of typical pectus excavatum morphology. However, its frequency has not been reported, which is the object of this study. We also hoped to facilitate the genetic study by classifying those variants that did not occur in our clinical practice frequently enough to be able to present them reliably.

A random sample of 300 pectus excavatum patients, treated surgically at the Children's Hospital of The King's Daughters Eastern Virginia Medical School, Norfolk, VA, U.S.A. (CHKD/EVMS, U.S.A.) was studied and classified. The study was approved by the Eastern Virginia Medical School internal ethical review board (number 01-05-EX-0175). Three hundred patients were selected to ensure that the study could be evaluated in detail within an acceptable time frame, and to have a group that was large enough to encompass the variety of morphologies observed in the cohort of over 2300 patients in this unit. Patients with Ehlers–Danlos, Marfan or other genetic syndromes were excluded from the study, as were patients with complications due to comorbidities, such as congenital heart disease or congenital diaphragmatic hernias, and those with a previous history of thoracic surgery. There was no attempt to correlate other pathologies, the frequency or time evolution of a particular morphology.

Medical records, preoperative photographs and computed tomography (CT) images of 300 patients of both sexes with identified pectus excavatum were examined retrospectively, with special attention paid to the defining characteristics identified in our previous study: Haller index, asymmetry index, relative length of the depression with respect to the sternum, fraction of sternum affected, degree of sternal torsion and localized (cup-shaped), diffuse (saucer-shaped), trench-like or Currarino morphology.4 A total of 17 CT scans were considered unsuitable, leaving 283 patients for the study.

Eighty percent of patients were male. They were aged between 4 and 30 years at the time of the CT scan; the median age was 14.5 years. Minimally invasive repair of the pectus excavatum was performed in all the cases (Nuss procedure). Two staff surgeons (DN and RK) separately evaluated the preoperative photographs taken at the clinic and CT scans of each patient. In the photographs they examined: the shape of the deformity (cup-shaped, saucer-shaped, trench-like or Currarino),8 the position of the center of the depression (to the right, center or left of the midline), the upper, mid- or lower sternum, the length of the depression expressed as a fraction of the apparent sternal length, the symmetry of the depression (do the sides of the depression have a similar slope, or does one side have a steep slope and the other a slight, long slope?) and the presence of a winged or bell-shaped thorax (anterior protrusion of the inferior costal cartilage) (Fig. 1). These characteristics were chosen because they were significant morphological observations in a previous study of our large reference group.4

Fig. 1.

Information on the protocol used to show the morphology in the preoperative photograph of the patient. (A) Median/sternal line chest; (B) length of sternum affected; (C) area affected by depression; (D) vertical length of deformity. Note how the area is affected by the depression asymmetry to the right.

(0.1MB).

On the CT images, the deepest point of the depression, both in the craniocaudal and sagittal plane, symmetry (comparing the slope of both sides), length of the craniocaudal depression, sternal torsion (and whether this was greater or less than 30°), Haller index and asymmetry index were evaluated (Fig. 2). To this end, the methodology in our previous study was strictly followed.9 Most of the CT scans were obtained before digital imaging techniques became available, and the measurements were done manually. The radiologic assessment was not examined.

Fig. 2.

The CT scan shows the sternal torsion and chest measurements.

(0.11MB).

Since the majority of patients at CHKD/EVMS come from outside our area, most of them were selected initially based on their place of residence. Thus, we see a large number of patients with “typical” pectus excavatum.

Other rare variants such as the well-known laterolateral juxtaposition of pectus excavatum and pectus carinatum, were not included in the random sample or in the analysis.

Surgical decision-making is affected by the type of pectus excavatum deformity presented by the patient when performing the Nuss procedure. Two bars are required in patients with trench-like dysmorphology. In many of the patients with a larger saucer-shaped pectus excavatum, placement of a second bar is also useful. Surgeons who use open surgery or the Ravitch procedure may have to perform additional resections of the costal cartilage and move the position of the sternotomy.

Recent treatment of a large number of patients with pectus excavatum has provided the opportunity to study its mode of inheritance. Family history is common in this disorder.2 A study of the family trees of 34 families in which there were two affected siblings showed that there was autosomal dominant inheritance in 41% of families, autosomal recessive in 12%, X-linked in 18% and complex inheritance patterns in the remaining patients (29%).5

Horth et al.10 evaluated 48 family trees with a total of more than 2000 individuals. These authors found clear evidence indicative of autosomal recessive, genetic control for this disorder. The proportion of males to females in their patients was 3.8–1, very similar to the 4:1 observed in our group. They considered that it was likely that there was more than one mutation responsible for the disorder. Although this group was able to evaluate 10 clinical features often associated with pectus excavatum (e.g. thin, tall, and long fingers), they were unable to correlate these characteristics with the phenotype.

The correlation of the genetic findings with the phenotype may be obtained by knowing the frequency of the phenotypes. In order to determine this frequency, 283 of the 1215 patients who had been previously diagnosed with a severe type of pectus excavatum and had been treated with a minimally invasive repair were randomly selected. Preoperative photographs and CT images were available for all the patients, and were used to classify them by applying the method described in 2006.4

The study limitations derive from patient selection that was not population-based. While we recognize this fact, the classification of patients with a typical deformity (a deformity that is severe enough to warrant an operation) has the value of a clear diagnosis. The exclusion of patients with Marfan syndrome and Ehlers–Danlos syndrome was necessary due to the difficulty of the definition. Of the 1215 patients presented in 2010, Marfan syndrome was identified in 2.8%, but there were features that were clearly suggestive of Marfan syndrome in an additional 17.8%, although there was no genetic confirmation of the diagnosis at the time of the preoperative evalation.3 Patients with Ehlers–Danlos syndrome were even less common.

The most common type of pectus excavatum is cup-shaped, symmetrical, to the right of the midline, and involves the lower sternum. However, other phenotypes like the long trench, which extends much further in the cranial direction, were observed with significant frequency. Although photographs enable its shape to be examined, definition of the deformity is more reliable by CT scan.

The authors declare that they have no conflict of interests.