The method of CLE visualization
CLE navigation was done to confirm the presence of the ENS by comparing the CLE findings with histological findings. Our observation of the target segments confirmed the presence of abundant ganglion cells although the procedure was not a surgical navigation using CLE to detect the segment to be resected but rather a trial evaluating the technical feasibility of the procedure during conventional surgery for Hirschsprung’s disease and allied disorders. CV (Muto Pure Chemical Co., Ltd., Tokyo Japan) was injected from the serosal side into the muscular layer for visualization by CLE. The dye was injected locally until the serosal side of the intestines became grossly purple; then the intestines were immediately resected to minimize any possible adverse effect of the dye on the patients. The scanning time was kept to less than 30 minutes based on the average time required for a conventional intraoperative pathological diagnosis.
Cresyl violet (CV)
CV is an organic compound with the chemical formula C19H18ClN3O. It is a basic dye commonly used in histology. CV was able to provide the clearest image of the ENS1, 2 among several dyes tested, such as fluorescein, Acriflavine3, and NeuroTrace6. In addition, CV is apparently safe despite not being designed for use in the clinical setting. CV has been used widely in the histopathology lab for Nissl staining7 or as a lysosomal marker8. Moreover, it has been used clinically for chromoendoscopies in Japan since the 1980’s and has an established safety profile9-13. CV has an emission and excitation wavelength of 630 and 585 nm, respectively8, which the CLE device used in the present study can detect. Thus, CV was the best available dye for ENS visualization by CLE in the present study.
The ethics and consent for clinical use of CV staining
As CV is not authorized for clinical use, careful attention was paid to designing the method of delivery into the intestinal muscle layers in vivo. First, CV was injected immediately before intestinal resection. Prior to its injection, the blood vessels in the mesentery were ligated and dissected to decrease the risk of CV circulation. Second, the observation time was kept to less than 30 minutes to decrease the risk of adverse effects. Third, the areas of the intestines into which CV was injected were resected in toto immediately after observation to minimize the amount of residual CV.
The aim and method of the present study, the potential risk of adverse effects, and the right to refuse participation prior to surgery were explained to the patients’ guardians, who gave their informed consent before enrollment. No complications were observed during or after the procedure in either case.
All areas observed by CLE were pathologically evaluated, and the findings were compared against those of the CLE. All specimens were fixed with 10% formalin after surgical resection and sliced into horizontal sections. The ENS morphology observed in the pathological specimens stained with hematoxylin and eosin (HE) were then compared with the CLE findings.
GastroFlex-UHD (Mauna Kea Technologies, Paris, France) and CellVizio system (Mauna Kea Technologies, Paris, France) were used. The external diameter of the probe was 2.5 mm, the field-of-view of the obtained image was 240 μm in diameter, the imaging rate was 12 frames/s, and the image depth was 55-65μm.
The first case was that of a 2-year-old male patient with a past medical history of Lynn’s sphincteromyotomy for Hirschsprung’s disease at another hospital during the neonatal period. The chief complaint at his current presentation was constipation despite the previous operation. An enema revealed a stenotic region in the rectum. The pathological results of a rectal mucosal suction biopsy revealed the absence of ganglion cells with nerve fiber hypertrophy. The recto-anal inhibitory reflex (RAIR) on rectal manometry using a dilating balloon was also negative. Based on these findings, the short-segment type of Hirschsprung’s disease was diagnosed, and the endorectal pull-through procedure using the prolapsing technique was performed to resect the aganglionic segment. A laparotomy from the umbilical approach was performed, and an intraoperative pathological biopsy specimen of the intestine on the oral side 5 cm from the stenotic region in the rectum revealed abundant ganglion cells on later histopathological analysis of the frozen sections. Then, trans-serosal CLE observation was performed just above the biopsied region of the colon after a transanal mucosectomy immediately prior to the colon resection. (Fig 1)
The second case was that of a 1-year-5-month old female patient who received the diagnosis of allied disorder of Hirschsprung’s disease during the neonatal period. She was admitted to the hospital due to repeated, bilious vomiting on the day of birth. An ileostomy was placed on day 8, and a full-thickness biopsy of the ileum revealed an allied disorder of Hirschsprung’s disease. The pathology of the segment showed a markedly thickened muscular layer and solitary ganglion cells in the mucosa and intramuscular layer. Additionally, ganglion cells were observed in the myenteric plexus. Based on these findings, an unclassified type of allied disorder of Hirschsprung’s disease was diagnosed. As intestinal motility improved with the patient’s development, a Bishop-Koop type enterostomy was placed. Defecation was achieved, and stoma closure was planned. Trans-serosal CLE observation was performed near the stoma on the anal side just before the resection and anastomosis (Fig2).