Clinical and Endoscopic Characteristics of Chinese Cronkhite-Canada Syndrome Patients: A Retrospective Study of 103 Cases

Introduction: Cronkhite-Canada syndrome (CCS) is a rare non-inherited disease characterized by extensive gastrointestinal (GI) polyposis and ectodermal dysplasia. So far, most of CCS related literatures are published as single case report or reviewed with limited case numbers. Our study was to update the clinical and endoscopic characteristics of Chinese CCS patients. Methods: This retrospective study was conducted in 103 Chinese CCS patients (102 cases from literatures and 1 case from our department). Their clinical and endoscopic data were collected, and statistical analyses were performed. Results: (1) In Chinese population, people aged 50–70 years (62.62%) had a high incidence of CCS, and the ratio of male-to-female was 2.68:1. (2) The diverse range of GI manifestations was observed in all the patients, and almost all the patients had at least 1 symptom of ectodermal dysplasias. (3) All CCS patients presented multiple polyps in the GI tract except esophagus, and the size and appearance of polyps were diverse. Congestion, edema, and erosion were very common on the surface of polyps (96.83%) and the surrounding mucosa (85.71%). (4) The common pathological features of polyps were hyperplastic polyps (49.25%) and tubular adenomatous polyps (44.78%). The prevalence of cancer was 5.97% in Chinese CCS patients. Conclusions: Middle-aged and elderly people are the high-risk group. Various GI symptoms are observed in Chinese patients; the typical endoscopic finding is multiple small sessile polyps. These GI polyps have a chance of malignant potential. Long-term endoscopic surveillance and follow-up are recommended for the Chinese CCS patients.


Introduction
Cronkhite-Canada syndrome (CCS), also known as a multiple gastrointestinal (GI) polyps-hyperpigmentation-alopecia-onychodystrophy syndrome, was first reported by Cronkhite and Canada in 1955 [1]. CCS is a very rare non-inherited disease. The highest incidence of CCS is reported in Japan as approximately 0.00037% [2]. Up to date, there are over 500 cases all over the world. The pathogenesis of CCS still remains unclear. Most scholars believed that CCS was an autoimmune-related disease, especially closely related to IgG4 [3,4].
GI disorder and ectodermal dysplasia are the most common symptoms presented in CCS patients. Usually, at least 1 ectodermal dysplasia symptom is shown in CCS patient such as alopecia, onychodystrophy, and skin hyperpigmentation. A wild range of GI symptoms is observed including abdominal pain, diarrhea, hypogeusia, and weight loss as the initial symptom. According to the initial symptoms, patients could be divided into 5 types: diarrhea (type1), hypogeusia (type2), xerostomia or a strange sensation in oral cavity (type3), abdominal pain (type4), and alopecia (type5) [5]. Endoscopic examination is the most important in all of the auxiliary examinations. It is manifested as diffuse multiple polyps in the GI tract, most of which are sessile polyps with various shapes. Currently, CCS diagnosis is mainly relied on typical clinical manifestations and endoscopic findings. At present, there is no effective specific treatment for CCS. The most used approaches include symptomatic remission, nutritional therapy, and corticosteroid. The prognosis of CCS is very poor; 5-year survival rate is <45% [6].
So far, the majority of CCS related literature are published as single case report or reviewed with limited case numbers. The study with the maximum number is 210 cases from Japan [7]. It mainly explored the best potential of treatment and monitoring strategies by evaluating the treatment efficacy of endoscopy and other approaches. As for Chinese CCS patients, the most recent literature in international journal was published in 2013 with limited cases (49 Chinese CCS patients) [8]. It summarized the clinical characteristics of CCS patients to better understand CCS in China. Up to date, there is still very little description and analysis of CCS' features, especially the specific endoscopic findings in Chinese patients. Our study is to review all the documented Chinese CCS cases so far and update the clinical and endoscopic manifestations in Chinese patients.

Materials and Methods
A literature search of the CBM Web (year range: 1978-March 2020), WanFang Data (1989-March 2020), and PubMed (1950-March 2020) was performed to identify publications with "Cronkhite-Canada"or "Cronkhite-Canada case report" in the title or abstract. Inclusion criteria are as follows: all patients were Chinese and had complete endoscopic reports; exclusion criteria are as follows: confirmed duplicate cases, suspected duplicate cases, and cases with the missing of endoscopic results. All identified cases were collected. 102 cases were reported in the literature, and 1 unpublished case from our department was added. Thus, this retrospective study encompassed a total of 103 patients.
The SPSS24.0 software was utilized for statistical analyses. The χ 2 test was used to calculate the statistical significance between groups. If the theoretical frequency was <1, or 1 ≤theoretical frequency <5 in R × C table was >1/5 of the total numbers, statistical significance was determined with the Fisher's exact test. p values <0.05 were considered statistically significant.

Sex and Age Distribution
Of 103 Chinese CCS patients, 75 were men and 28 were women, which yielded a male-to-female ratio of 2.68:1 (shown in Fig. 1a). The age range of patients was from 9 to 82 years old, and the majority of groups were between 50 and 60 years old (30.30%) and 60 and 70 years old (32.32%). The average age was 57.39 years old (shown in Fig. 1b).

Clinical Manifestations
GI discomfort was the initial symptom and the reason of seeking for medical treatment for most CCS patients.  Fig. 2a). The edema was often observed in lower limbs (33/38). Patients could lose 10 kg a month in some severe cases. Almost all the patients had at least 1 symptom of ectodermal dysplasia: alopecia (91/95), hyperpigmentation (90/94), and onychodystrophy (101/101) (shown in Fig. 2b). Scalp hair loss occurred first, followed by the loss of eyebrows, eyelashes, and the hair of limbs. Different sizes of tan spots were distributed on the palms, soles, face, neck, limbs, and oral with a clear boundary. Nails are characterized by tarnish, brittle, and cracking. Some patients had completely chipped nails.

Laboratory Examinations
A peripheral blood test mainly showed anemia and low serum protein. Decreased hemoglobin was shown in 38 patients. Most of them had mild and moderate anemia, and only 2 patients had severe anemia with <30 g/L hemoglobin. Other indexes of routine blood test were not significantly abnormal. Serum total proteins were low in 55 of 63 patients, and serum albumin levels were low in 72 of 84 patients, which indicate malnutrition in many patients. Among the 76 patients with fecal test records, the occult blood test was positive for 77.63% (59/76). Only 24 patients had the results of Helicobacter pylori (HP) test, 3 of them were tested positive (shown in Fig. 2c).

Endoscopy and Pathology
All 103 patients underwent esophagogastroduodenal endoscopy and/or colonoscopy. However, the author only presented the result that they considered important; the effective number of cases for endoscopic features was not all 103. All the endoscopic features were summarized in Table 1, and the variable appearances of gastric, small intestinal, and colonic polyps in our department were shown in Figure 3.

Location
The polyps could be found all over the digestive track: 92.23% (95/103) in stomach, 73.79% (76/103) in small intestine, and 94.17% (97/103) in colon and rectum. They were mainly located in the gastric antrum (36/103) and duodenal bulb (39/103). The esophagus was rarely involved (4/103).    Size Forty-four of 103 patients had the detailed information about polyps' size. The size was varied from ≤0.5 to >2.0 cm. By comparing polyps among stomach, small intestine, colon, and rectum, the whole GI tract was dominated by small polyps with diameters ≤0.5 cm (88.64%). The polyps >2.0 cm was mentioned in 16 patients. There was no statistically significant difference in the size of polyps in comparison of various parts of GI tract (p = 0.794 > 0.05) (shown in Table 1).

Basement
Of 103 patients, 41 had the description of basement features. Pedicle, sub-pedicle, and sessile polyps were accounted for 46.34, 34.15, and 90.24%, respectively. In comparison of stomach, small intestine, colon, and rectum, there was no statistically significant difference in the basement feature of polyps (p = 0.559 > 0.05) (shown in Table 1).

Shape
The polyps were different in shape according to the 43 patients. 11 different morphologies of polyps were described in total: nodular, lobulated, strawberry-like, coral-like, cauliflower-like, granular, grape-like, herpes-like, beads-like, circular, and hemispherical. Nodular polyps (25/43) were the most common, and other shapes were scattered. There was no statistically significant difference in the shape of polyps located in different parts of GI tract (p = 0.749 > 0.05) (shown in Table 1).

Surface
According the records from 63 patients, the polyps' surface presented a variety of features such as smooth; reddish, engorged, edema, and erosion; strawberry-like change; lobulated change; spotted change; and cauliflower-like change. Reddish, engorged, edema, and erosion were most commonly observed in 61 patients (61/63). There was also no statistically significant difference in the feature of polyps' surface in different parts of GI tract (p = 0.764 > 0.05) (shown in Table 1).

Surrounding Mucosal Lesion
In order to better understand the characteristics of CCS, we further investigated the surrounding mucosa of polyps. Its detailed description was found in 49 of 103 cases. The surrounding mucosa could be smooth; reddish, congestion, edema, and erosion; and pale and pigmentation under the endoscopy. Reddish, congestion, edema, and erosion in adjacent mucosa were observed in almost 42 (85.71%) patients. Only 5 patients reported smooth. The different levels of inflammation were observed in nearly all the mucosa surrounded polyps. And no statistically significant difference found in mucosal lesions (p = 0.199 > 0.05) (shown in Table 1).

Pathology
Neoplastic and non-neoplastic polyps were both visible in CCS patients; 5.97% of the patients were cancerous (4/67). Hyperplastic polyps and tubular adenoma were the most common ones, accounting for 49.25 and 44.78% respectively. Furthermore, hyperplastic polyps were most common in the stomach (19/33), and tubular adenomas were most common in the colon (17/30) (shown in Fig. 4).

Treatment and Prognosis
Medical treatments reported in the literature vary. Some patients may use one or more of these treatments, and almost every patient was given symptomatic remission and nutritional support treatment. 50 patients (51.02%) received corticosteroid, which was used the most. Based on the view of some scholars that CCS was

Discussion
CCS is a rare disease characterized with generalized GI polyps and ectodermal dysplasia. The etiology and pathogenesis are not clear. There is no effective treatment available now. In this study, we reviewed clinical information from all documented Chinese CCS patients so far, especially the detailed description of endoscopic characteristics, so as to update and better understand CCS in Chinese patients.
Our data have revealed that people aged 50-70 years had a high incidence of CCS in China, and most of them were male (72.82%). The previous study with large sample size has reported that males and people aged 50-60 were the high-risk group [5][6][7]. There were also a few sporadic cases reported among young people [9,10]. Our findings from Chinese CCS patients were very similar with those results from other areas. We also noticed that Beijing was the city where the most patients were located compared with other places in China (26/103). However, the location in patient record usually was the place where patients received the treatment. Since the healthcare system in Beijing is much stronger than most areas of China, lots of patients from other places would come to Beijing for medical treatment. Tokyo, Aichi, Osaka, and Hyogo in Japan had the highest incidences, which might be proportional to the relative population density [5]. Therefore, this result of location statistics is still open to question.
CCS is believed to be a non-genetic disease. Till now, only 2 family cases have been reported and most of the cases are sporadic. However, both the mother and the child have APC gene c.3921-3925delAAAAG (p.Ile1307fsX6) mutation [11,12]. What's more, genetic sequencing analysis also found that PRKDC and MUC3A mutation might contribute to CCS pathogenesis in other case report [13,14]. These findings suggested that CCS might not be completely non-hereditary, and further investigation of genetic mutations in more CCS patients remains to be done. The pathogenesis of CCS is still unclear. Many researchers believed that CCS was an autoimmune disease because that IgG4-positive cell infiltration was observed in polyps and corticosteroid or azathioprine treatment was effective in some cases [3,4]. However, the study from Li et al. [15] showed that CCS was not the variant of IgG4-related disease although the IgG4-positive plasma cells might be involved in CCS pathogenesis. A study stated that CCS might be related to allergies [16]. In addition, Goto [5] reported that mental stress and physical fatigue were the important risk factors for CCS. HP might also be another factor since eradication of HP could reduce some symptoms in CCS patients [17,18]. There is very limited information in our study. Only 2 patients had IgG4 plasma cells positive in polyps detected by immunohistochemistry. And 3 out of 24 patients were tested positive for HP. Further investigations are needed to address these questions throughout.
GI disorder is the major clinical manifestations of CCS, which is also the initial symptom of most patients, and for some patients in the early stage of CCS, it is difficult to distinguish from inflammatory bowel disease, intestinal tuberculosis, and other diseases [19,20]. The similar GI symptoms were observed in all 103 Chinese patients including abdominal pain, diarrhea, and so on. Hypogeusia was found in some patients. Some studies have shown that hypogeusia could be caused by inflammation, infection, or other oral mucosa lesion, and the deficiency of zinc and copper could also lead to hypogeusia [21]. Serum electrolyte disturbances are very common in CCS patients due to malabsorption and GI loss [8]. However, we did not get enough data to confirm the connection between hypogeusia and electrolyte disturbances in current study. Ectodermal dysplasia is the other major clinical manifestations of CCS, and some patients only have ectodermal dysplasia symptoms [22,23]. Almost all patients in our study presented with at least 1 symptom of ectodermal dysplasia. The underlying mechanism is still unknown. Hair loss might be caused by malnutrition [23], but Horikawa et al. [24] has proposed that it was at least partly due to the autoimmuntiy, and Chuamanochan et al. [25] has reported that nail change in CCS patients was the result from inflammatory response, not only the malnutrition.
Previous studies have shown that polyps in CCS patients were diffusely distributed throughout the entire alimentary tract except esophagus [6,8]. However, a Jap-DOI: 10.1159/000514354 anese study showed that 12.3% (26/210) of patients had esophageal polyps [7]. In our study, polyps were observed along the gastrointestinal tract by endoscopy, and only 4 patients had polyps in esophagus. Most polyps were sessile or sub-pedicle, varying in size from a few millimetres to 1.5 cm in diameter. They had various endoscopic appearances, accompanied by reddish and edema on the surface of the polyps [7,8,26,27]. Further observation of the small intestine by capsule or double balloon endoscopy revealed excessively elongated or atrophic intestinal villi [28][29][30]. Likewise, we found that the polyps were multiple sessile small polyps, and the surface of polyps and surrounding mucosa were presented with reddish, engorged, edema, or erosion. Although polyps were diverse in shape, nodule-like shape was the most observed. These endoscopy findings could help to better distinguish CCS from a number of polyposis syndromes such as Familial adenomatous polyposis, Gardner syndrome, Turcot syndrome, Peutz-Jeghers syndrome, Juvenile polyposis syndrome, and Cowden syndrome.
Initially, CCS had been considered to be a hamartomatous polyposis syndrome [21]. But it has been realized that the polyps of CCS could be inflammatory, adenomatous, and hyperplastic now. Based on the study of 14 consecutive CCS patients registered at the Mayo Clinic, the incidence of adenomatous polyps had reached 71% [4]. Our data showed that hyperplastic polyps and tubular adenomas were the most common polyps in Chinese CCS patients. The reason for this difference needs to be investigated and researched by a larger sample size. Yashiro et al. [31] have confirmed genetic changes linking sawtooth adenoma to colorectal cancer and proposed a serrated adenoma-carcinoma sequence in CCS. The data from Haghighi et al. [32] have reported that the incidence of cancer in CCS was 15.6% (78/500). It is still controversial whether cancer is a coincidence in CCS patients or developed from benign polyps. In general, the overall risk of cancer in CCS is 5-25% [33][34][35]. Our study has found that the prevalence of cancer was 5.97% in Chinese CCS patients. This rate in Chinese patients is much lower than others in the world. But it is higher than the national cancer incidence rate of 0.2781% announced by National Central Cancer Registry of China (NCCRC) in 2014 [36]. Therefore, long-term endoscopic surveillance and follow-up are recommended for Chinese CCS patients.
Because of the unclarified etiology of CCS, there is no consensus on the CCS management. At present, the treatment includes symptomatic remission and nutritional therapy, corticosteroid, antibiotic, non-steroidal anti-in-flammatory drugs, acid inhibitor, immunomodulator, Chinese herbal medicines, therapeutic endoscopy, and surgery. Study has reported that corticosteroid and immunomodulator were the most effective treatment so far, which could maintain clinical remission for >8 years [37,38]. In our study, almost every patient was given symptomatic remission and nutritional therapy. 50 patients (51.02%) received corticosteroids. A total of 68 patients were followed up, 12 (17.65%) patients had died of the disease, and 45 (66.18%) patients had remission. The cause of death is mainly the disease itself or the complications: severe cachexia, congestive heart failure, embolism, sepsis, and shock. Compared to the previously reported 55% mortality rate for CCS, the prognosis of CCS is much better than before.
This retrospective study is with the largest case number including all the documented Chinese CCS patients so far. However, there are still some limitations to this study. All the data were collected from the case report, and some of the data were incomplete, especially the laboratory test, pathology, and prognosis. The scarcity of cases limits the understanding of the disease. It will be improved by the future investigation.
The polyps in Chinese CCS patients are multiple small polyps along GI tract, and there are various shapes. And they are usually accompanied by hyperemia, edema, and erosion on the surface of polyps and surrounding mucosa. There is no significant difference in characteristics of polyps under endoscopy between Chinese and foreign CCS patients. There is a chance of malignant potential for Chinese CCS patients; long-term endoscopic surveillance and follow-up are recommended.

Statement of Ethics
Our study protocol complied with the ethical guidelines of the Declaration of Helsinki and was approved by the institutional ethics committee of Zhongnan Hospital of Wuhan University (No. 2020065K), and waiver of informed consent was approved as the bulk data were received and analyzed anonymously.