Abstract Background/Purpose: The cause of the pectus excavatum (PE) remains unclear, although some results of research have indicated that the disturbance of the sternum or costal cartilage might be responsible for this deformity. But no decisive evidence has been gained. The authors have analyzed the biomechanical, morphologic, and histochemical properties of the cartilage in PE and intend to support the belief that the disturbance of the cartilage might contribute to the development of PE. Methods: Thirty-eight specimens of the sixth cartilage were obtained at operation for the PE group (aged from 3 to 6 years; mean, 4.2 years). And 28 specimens of the control group (aged from 3 to 6 years; mean, 4.4 years) were gained from routine postmortem examinations in which the cause of death was unlikely to have affected the cartilage. The biomechanical test was carried out in a material testing machine (Shimadzu AG-10TA, Tokyo, Japan). The relation curve of load-deformation in tensile and compressive tests and the curve of load-time in the flexuous test were recorded automatically. The values of the ultimate strength and strain were calculated from this relation curve. The specimens also underwent H&E staining. The values of the area, circumference, mean diameter, maximal diameter, and morphologic factor of the cell and the nucleus of the cartilage in superficial and deep area were determined with the help of image analysis software (GT-2 model, China). The superficial zone (SZ) and deep zone (DZ) of the cartilage were examinated with electron microscopy (JEM-100SX, Japan). The distribution and intensity of type II collagen was shown by immunohistochemistry staining and analyzed with the image analysis software (GT-2 model, Huakang Co, Chengdu, China). The extent and distribution of proteoglycan were analyzed after Safranin-O and periodic acid shiff (PAS) staining. Results: The mean strength of the costal cartilage in the experimental group was less than that in the control group in terms of tension, compression, and flexure (P < .05). The shape of the stress-strain curve for tension and compression in the experimental group was different from the control group. The fracture load in the experimental group was less than in the control group in tension (1.5 MPa versus 2.8 MPa) and in compression (.2 MPa versus 8.3 MPa). The time of fracture in experimental group was 30 seconds compared with 38 seconds in control group. No denaturation or necrosis could be found in light microscopical examination. There was no manifestation of hyperplasia or hypoplasia in the costal cartilage of the PE group. In SZ and DZ areas, the pattern and the number of mitochondria, endoplasmic reticulum, and Golgi in the experimental group were the same as the control group in transmission electron microscopy. Furthermore, the distribution and the number of proteoglycan in the 2 groups did not show a significant difference both in SZ and DZ areas. Although the distribution of the collagen in SZ areas was normal, this pattern was disturbed in DZ areas in the experiment group. The results of type II collagen immunohistochemistry examination was concordant with that change. No significant difference between control and experimental group could be seen in Safranin-O and PAS staining for proteoglycan. Conclusions: The biomechanical stability of the cartilage was decreased in the PE group. This might be caused by the disorderly arrangement and distribution of the collagen in the cartilage of PE patients. J Pediatr Surg 36:1770-1776. Copyright © 2001 by W.B. Saunders Company.