Optical coherence tomography (OCT) serves as a non-invasive, real-time, high-resolution in vivo imaging technology, while optical coherence microscopy (OCM) is a product of the combination of OCT and confocal microscopy, compared with OCT, OCM has a richer source of contrast and higher spatial resolution than OCT[17]. In 1997, OCT technology was first applied to the diagnosis and treatment of digestive tract diseases[18,19]. Studies related to the application of OCT technology in the human digestive tract have been reported in all sites such as esophagus, stomach, colon, bile duct and pancreatic [20-24]. At present, the research on this technology in the digestive system is progressing rapidly, especially through the combination of digestive endoscopy, puncture needle, fine needle, catheter and other instruments. According to the real-time OCT imaging features, it can accurately evaluate the mucosa, proper layer, mucosa, muscle layer and part of the submucosa, clearly determine the structural characteristics and properties of the disease, and has great potential in the early screening and diagnosis of gastrointestinal tumors; and shows great advantages in detecting the surgical margin of tumor lesions. Moreover, Hariri et a[25]. have combined laparoscopy with OCT for the first time and successfully applied it intraoperatively, suggesting the feasibility of intraoperative laparoscopic OCT for real-time detection of surgical margins and lymph nodes to determine the appropriate extent of bowel resection and lymph node dissection. These studies fully demonstrated the advantages of OCT technology in disease screening, diagnosis, and surgical resection scope.
Hirschsprung’s disease (HSCR), one of the common congenital digestive tract malformations in pediatric surgery, has multiple diagnosis and treatment methods, while the difficulty in confirming the diagnosis preoperatively still exists, especially in newborns, when the ganglion cells are immature, and the complexity of histological manifestations often leads to the difficulty in judgment[26]. At present, the clinical methods for HSCR diagnostic examination has commonly used, including BE, ultrasound, anorectal pressure measurement, rectal mucosa biopsy, etc. BE examination has some false positives and false negatives (about 20%), the diagnosis cannot be affirmed or excluded, and only helps provide the approximate location of the migrated segment for the reference of choosing the surgical access (transanal or laparoscopic aid) [27], especially in the neonatal period, ultra-short and short segment due to barium stimulation, easily misdiagnosis or miss[28]. This was also confirmed in this study, which showed that 15 cases of negative preoperative BE and 18 cases of short segment type were diagnosed as common type HSCR after surgery. Confirming the diagnosis of HSCR still depends on rectal tissue biopsy (rectal mucosa or the whole layer). However, rectal mucosal biopsy has some disadvantages such as blindness, difficulty in standardizing the site, thickness and number of specimens in different cases, and the possibility of false negatives and inability to determine the extent of lesions in neonates, especially in preterm and low birth weight infants. The results of this study show that the OCT/OCM non-invasive diagnostic technology shows great potential in assisting HSCR diagnosis, pathological classification, and clarifying the scope of surgical lesions.
Westphal et al[29].obtained real-time endoscopic OCT images of fresh specimens of ileum, colon and rectum in a clinical trial, where the structures of mucosal and submucosal, glands, vessels, small pits, villi and crypts were observed. The results concluded that images provided by real-time endoscopic OCT were correlated with the tissue structure of gastrointestinal mucosa and submucosa. The visual stratified structure of the intestinal wall obtained by OCT/OCM scanning of human colon tissue was also confirmed in this study. Xiong, HL et al[30]. used OCT for the first time in SD rats to examine colon tissue without HSCR and with HSCR treatment for 3 and 6 weeks by establishing the HSCR Sprague-Dawley (SD) rat model and recording the change of colon tissue attenuation coefficient and myometrial thickness difference. The study initially demonstrated the feasibility of OCT for HSCR imaging, showing OCT in HSCR suspicious tissue in vivo diagnosis, and positioning targeted biopsy and invasive surgical treatment.
This study differed from the Xiong, HL[30], in that it exploits the principle of different reflectance intensities of different tissue structures and the high-resolution feature of OCM to detect human in vitro HSCR tissues for the first time, and observes and summarizes the OCM images of human HSCR tissues and the corresponding histological image features, and initially explores the feasibility of OCM for human HSCR imaging. According to previous studies, the average thickness of the colon tissue of healthy adults is 2.2-2.3 mm, and that of children aged 0–4.99 years is approximately 1 mm. In the OCM images, there were significant differences between the normal colon tissue and the HSCR, and also between the OCM images of the dilated and narrow segments (Fig. 4). The muscle layer in HSCR had higher attenuation, and the tissue light scattering was significantly reduced in the dilated segment compared to the stenotic segment, which may be caused by hypertrophy, cytoskeletal changes, vacuolization of muscle tissue, and reduced myofilaments of muscle cell clusters[30]. The thickness of the muscle layer in the dilated segment in HSCR was significantly increased, which was caused by hypertrophy of muscle tissue.
The data of the present study showed that the OCM can identify the abnormal segmental intestine tube due to barium stimulation, which results in a more accurate diagnosis of HSCR and a clear classification. The diagnostic sensitivity of OCM was 92.66%, specificity 90.00%, AUC=0.913; the diagnostic sensitivity of BE was 72.48%, AUC=0.862; the sensitivity and AUC of OCM examination were higher than those of BE examination, and the difference between the two sensitivities was statistically significant (χ2=16.962, P<0.001), indicating that OCM examination can reduce the rate of missed diagnosis, as a supplement to BE examination; the Kappa values of OCM and BE examination were 0.627, and 0.307, respectively, indicating that the consistency between OCM and histopathological examination was significantly higher than that of BE examination.
Our study has several deficiencies: (1) No specific changes in the attenuation coefficient of colonic tissues were detected, and no criteria for determination of myometrial thickness were developed. (2) Sample selection was not comprehensive enough, in which the HSCR type was incomplete; and whether there were differences in the sensitivity and specificity of OCM for the detection of different types of HSCR; secondly, OCM sensitivity for HSCR in different ages needs to be further explored. (3) In this study, the detection of ex-vivo specimens by OCM did not reflect the advantages of real-time OCT in-vivo detection. In the future, OCT and colonoscopy can be combined to further investigate HSCR detection and diagnosis. It requires future multicenter studies with large samples to develop appropriate diagnostic criteria based on the variation of HSCR myometrial thickness in different age groups and the combination of algorithms to calculate the attenuation coefficient.