Abbreviation
INTRODUCTION
CD is a chronic inflammatory disorder that arises from interactions among genetic susceptibility, environmental exposures, the intestinal microbiota, and dysregulated mucosal immunity.1,2) The higher concordance of CD in monozygotic twins than in dizygotic twins supports a genetic contribution, whereas incomplete concordance indicates that genetic background alone does not determine disease development or phenotype.3,4) Reports of concordant CD in presumed monozygotic twins remain uncommon in East Asia. We report twin brothers who developed ileocolonic and perianal CD 2 years apart and required different surgical interventions. Rather than attributing differences in their early clinical courses to specific genetic or environmental factors, we focus on the practical surgical lessons: prompt investigation of a symptomatic cotwin and phenotype-based selection of treatment for each patient.
CASE PRESENTATION
The pedigree is shown in Fig. 1. The patients and their family reported that the brothers were monozygotic twins; however, zygosity testing was not performed.
No other family member shown in the pedigree had been diagnosed with inflammatory bowel disease. Both brothers were nonsmokers and students. They lived in the same family home, had no substantial differences in diet, shared the same childhood living environment, and did not routinely use antibiotics.
Before their CD-related presentations, neither brother had a notable history of infection.
Twin 1
The elder brother had no relevant medical history when, at the age of 16 years, he underwent incision and drainage of a perianal abscess at another hospital in April 20XX. In July, he was admitted elsewhere because of abdominal pain and hematochezia. Colonoscopy showed continuous severe inflammation extending from the rectum to the descending colon, prompting suspicion of CD.
Corticosteroid treatment was ineffective. Stool culture yielded Salmonella species, and his symptoms temporarily improved with antimicrobial treatment.
In August, repeat colonoscopy showed extensive mucosal necrosis extending from the rectum to the sigmoid colon and ulcer scars in the ileocecal region (Fig. 2A). Epithelioid granulomas were identified in an ileocecal biopsy specimen, and stool culture was negative for Salmonella.
After discharge, abdominal pain and hematochezia recurred. Colonoscopy in September showed severe rectosigmoid stenosis; stool culture again yielded Salmonella, and antimicrobial therapy produced transient improvement. He nevertheless required repeated admissions for recurrent abdominal pain, diarrhea, hematochezia, and vomiting.
CT showed rectosigmoid stenosis with proximal inflammation. Double-balloon enteroscopy demonstrated longitudinal ulcer scars in the small bowel (Fig. 2B).
By November, the rectal stenosis was impassable with the colonoscope (Fig. 2C) and was confirmed by contrast enema (Fig. 2D).
He was transferred to our hospital after 5 admissions.
On admission, his height was 170 cm and his weight was 51 kg, representing a 9-kg weight loss. His abdomen was flat and soft. Laboratory tests showed a hemoglobin level of 12.2 g/dL and a C-reactive protein level of 0.53 mg/dL; the serum albumin level was 4.6 g/dL.
The combined clinical, endoscopic, and histological findings supported ileocolonic CD with stricturing behavior and perianal disease (Montreal classification: L3, B2p).
Because recurrent obstructive symptoms were attributable to fixed rectosigmoid stenosis, surgery was undertaken. The terminal ileum was adherent to the rectosigmoid colon and was carefully separated. Dense retroperitoneal adhesions necessitated conversion to an open surgery.
An anterior resection with a double-stapled colorectal anastomosis was performed. Histopathological examination showed scattered epithelioid granulomas (Fig. 3).
The postoperative course was uneventful, and he was discharged on POD 8. Vedolizumab was initiated after discharge. At more than 2 years of follow-up, he remained free of recurrent obstructive symptoms.
Twin 2
2 years later, at the age of 18 years, the younger brother developed anal pain and bleeding. A perianal abscess was incised and drained at another hospital. Because his cotwin was being treated for CD, he was referred promptly to our hospital.
His height was 172 cm and his weight was 64 kg. Laboratory tests showed a white blood cell count of 10300/µL and a C-reactive protein level of 2.78 mg/dL; the serum albumin was 4.0 g/dL, and the hemoglobin level was 14.4 g/dL.
Colonoscopy showed longitudinal ulcers extending from the terminal ileum to the cecum (Fig. 4A) and scattered aphthous ulcers throughout the colon and rectum (Fig. 4B).
Anorectal examination demonstrated a low intersphincteric fistula with a secondary opening at the 1-o’clock position (Fig. 5A). These findings supported ileocolonic CD with non-stricturing, non-penetrating behavior and perianal disease at presentation (Montreal classification: L3, B1p).
Because the fistula remained painful, an examination under anesthesia was performed. Hydrogen peroxide injected through the secondary opening exited from the primary opening at the 12-o’clock position, confirming a straight tract. The tract was curetted, and loose seton drainage was established with a Penrose drain and 2 vessel loops (Fig.
5B). Curetted tissue contained epithelioid granulomas (Fig. 6).
He was discharged on POD 4, and vedolizumab was initiated after discharge.
3 months after surgery, all tubes were removed. 6 months after surgery, the perianal fistula recurred with purulent drainage; therefore, vedolizumab was switched to adalimumab. 1 month after the switch, the perianal fistula improved, and the drainage ceased.
He required no additional surgery. Although this response was favorable, 7 months of postoperative observation remained insufficient to determine his long-term luminal or perianal disease course.
Selected published reports of concordant CD in twin pairs from Japan and other countries, with attention to surgical interventions, are summarized in Table 1.
DISCUSSION
The principal educational feature of this report is that presumed monozygotic twin brothers both required surgical treatment for CD, but for different immediate indications. Twin 1 required bowel resection for fixed rectosigmoid stenosis and recurrent obstructive symptoms, whereas Twin 2 required seton drainage for a symptomatic perianal fistula. After all drains were removed, fistula drainage recurred and resolved after vedolizumab was changed to adalimumab, underscoring the need for continued postoperative reassessment and timely modification of medical therapy.
Both brothers developed CD before the age of 20 years, with a 2-year interval between their onsets, ileocolonic and perianal involvement, and histological epithelioid granulomas. Nevertheless, treatment was determined by each patient’s contemporaneous anatomy and symptoms rather than by the cotwin’s phenotype.
An association between NOD2 variants and susceptibility to CD also supports a genetic contribution.5)
Population-based twin studies support a substantial genetic contribution to CD while demonstrating incomplete penetrance. In a recent Swedish nationwide cohort of 111080 twins, proband concordance for CD was 30% in monozygotic pairs and 2% in dizygotic pairs; the corresponding values for ulcerative colitis were 15% and 3%, respectively.6) The estimated heritability was 0.78 for CD and 0.57 for ulcerative colitis.6) These population estimates should not be interpreted as the probability that any particular symptomatic cotwin has CD, and they do not verify the zygosity of the present brothers. Earlier cohorts reported different absolute rates owing to study design and ascertainment.3,4)
Recent twin-based studies have investigated mechanisms beyond inherited sequence variants. A metagenomic study reported inflammatory bowel disease (IBD)-associated microbial signatures in healthy cotwins of affected patients.7) These associations do not identify the cause of disease in an individual pair. In the present patients, the histories of residence, diet, smoking, antibiotic exposure, and major infections were similar or unremarkable, but exposure data were not quantified, and no genomic, microbiome, epigenetic, or immune profiling was performed.
Thus, the cause of concordance or of the different operative indications cannot be inferred.
Previously published Japanese case reports describe 3 concordant twin pairs,8–10) and reports from Korea and Spain also document concordant pairs.11,12) Outside Japan, concordant twin pairs have often been examined within twin cohorts rather than presented as detailed individual surgical case reports. Consequently, reports with patient-level operative information suitable for direct comparison were limited in our targeted search. These examples are compared in Table 1; the selection is illustrative rather than a systematic global census.
In a Swedish-Danish series of 17 monozygotic pairs concordant for CD, 14 pairs developed the disease within 5 years of each other, and disease location was concordant in 11 pairs.13) The Spanish pair underwent complex bowel and duodenal operations for coloduodenal fistulas in both siblings.12) In contrast, our brothers required distinct operations at an early stage: resection for a fixed rectosigmoid stricture in 1 brother and drainage of a symptomatic perianal fistula in the other. The latter brother’s short follow-up precludes any claim of durable phenotypic divergence.
The surgical distinction is clinically relevant. In Twin 1, recurrent obstruction from an impassable, fixed rectosigmoid stricture despite medical treatment supported segmental resection; extensive adhesions required conversion to an open approach, and bowel continuity was restored with a colorectal anastomosis. In Twin 2, a painful low intersphincteric fistula without luminal stenosis was mapped under anesthesia, curetted, and drained with loose setons, thereby preserving the sphincter function.
These decisions addressed the immediate mechanical and septic problems while postoperative biologic treatment was adjusted according to subsequent symptoms. New gastrointestinal or perianal symptoms in a cotwin should prompt targeted evaluation, as occurred after Twin 2 developed a perianal abscess. This observation does not support routine endoscopy in asymptomatic relatives.
This report has several limitations. First, zygosity was based solely on information from the patients and their family and was not objectively confirmed; the brothers should therefore be regarded as presumed monozygotic twins. Second, Twin 2 was followed for only 7 months after surgery.
His classification as B1p describes his luminal phenotype at presentation and does not establish persistent divergence from Twin 1, because stricturing or penetrating complications may develop later. Third, although several relevant exposures were compared and appeared similar, the information was based on medical history and was not quantified; medication exposures other than routine antibiotic use were not systematically assessed. Consequently, the differences in surgical indications and postoperative courses cannot be attributed to nongenetic factors.
Finally, the initial isolation of Salmonella in Twin 1 complicated interpretation of his severe colitis; however, recurrent inflammation and stenosis after microbiological clearance, small-bowel ulcer scars, perianal disease, and epithelioid granulomas supported a diagnosis of CD.
CONCLUSIONS
Surgically treated ileocolonic and perianal CD developed 2 years apart in presumed monozygotic twin brothers. The known diagnosis in Twin 1 facilitated early gastrointestinal evaluation of Twin 2 when he developed a perianal abscess. The different immediate indications for anterior resection and seton drainage, followed by modification of biologic therapy after fistula recurrence in Twin 2, emphasize that management should be individualized and reassessed over time.
Objective zygosity testing, more detailed exposure assessment, and longer follow-up would be required before conclusions can be drawn about genetic concordance or divergence in disease behavior.