PMP is a rare disease with a large volume of extensively implanted gelatinous mucous substance on the surface of the peritoneum or omentum majus[1,2
]. The jelly-like substance is called pseudo-mucin, whose chemical properties are different from those of the mucous proteins. PMP was first discovered by Werth[1
] in 1884, and its incidence is approximately 2/10 000 in all the patients who are undergoing laparotomy. This disease, with a mean age of 60 years (range: 30-88 years) and a male:female ratio 1:3.4, is characterized by unrelenting pain of gradual onset, abdominal distension, and mucous ascites in literatures. Ultimately, the tumor may occupy the majority of the abdominal cavity and “jelly belly” syndrome may occur. Definitive clinical diagnosis is very difficult to make, and almost all cases are diagnosed with assistance of laparotomy. Detection of jelly-like ascites through abdominal puncture has high diagnostic value for this disease. Our patients with PMP were 24-76 years of age, (mean age, 52.5 years), and the female:male ratio was 1:1.2.
The causes of PMP have remained controversial for many years. It has been reported that the primary tumor of PMP is present in many organs, of which the most common are the appendix and ovaries, followed by the Fallopian tube, pancreas, and intestine. The primary foci are hard to detect in some cases[2,3
]. Most patients with PMP either suffer from appendiceal mucinous diseases (including cystadenoma and cystadenocarcinoma) or have a history of recent appendectomy[4-10
]. In women, PMP may be accompanied by ovarian mucinous tumors, which often occur bilaterally. If the tumor occurs unilaterally, it more often affects the right ovary[9,10
]. Clinically, about a third to a half of women with PMP have concurrent ovarian and appendiceal mucinous tumors.
Our hospital has treated 83 cases of PMP (45 men and 38 women) since the establishment of our hospital. The number of male patients was greater than female patients, which is different from that reported in the literature. Thirty-one of the 35 female patients in our study underwent appendectomy, and appendiceal mucinous tumors were seen in 28 of these. Three patients underwent surgery at other hospitals, and the pathological results of two were obtained. The pathological results of the remaining patient were unknown. For the two patients with pathological results, the pathological diagnosis was chronic inflammatory mass of the appendix in one patient, and no significant change in the appendix in the other. The remaining four patients did not have a history of appendectomy, and in three of them, the appendix was not explored during cytoreductive surgery for peritoneal tumor.
Twenty-four of the 35 patients had ovarian mucinous lesions, which were bilateral in 13 cases, on the right side in nine, and on the left side in two. Concurrent appendiceal and ovarian lesions occurred in 20 cases. It is controversial whether the ovarian lesions in these patients were primary or secondary to the appendiceal lesions.
In recent years, the techniques of immunohistochemistry and molecular biology have become more sophisticated, and have greatly enhanced our ability to identify the origin of this disease. In 1991, Young et al[5
] analyzed 22 cases of ovarian mucinous tumors and PMP-induced appendiceal mucinous tumors. Their results showed that PMP and ovarian lesions both originated from the appendix. Subsequently, Ferreira[6
] and Ronnett[7
] have made the same observation. Using immunohistochemical methods, Dong et al[10
] in China have analyzed CK7, CK20 and CA125 expression in peritoneal, ovarian and appendiceal tumors in women with PMP. These investigators drew the same conclusion. Szych et al[8
] have analyzed the k-ras
mutations and chromosome 18q, 17p, 5q and 6q alleles in patients with PMP. Their results support the conclusion that the ovarian lesions originate from the appendix in patients with PMP.
It has been reported in the literature that CK7 is a specific marker of primary ovarian epithelial tumors, which is rarely expressed in gastrointestinal epithelial cells[4,6
]; however, our experience has shown that CK7 is not specific enough, and sometimes CK7 is positive in typical gastrointestinal adenocarcinoma. CK20 is mainly expressed in the epithelial cells of the gastrointestinal tract[4,6
], but it can also be positive in intestinal-type ovarian mucinous tumors. MUC2 is a highly specific marker of intestinal goblet cells[4,6
]. MUC1 is often expressed in epithelial tumors of the breast and female reproductive systems[6
], but it is negative in epithelial cells of the gastrointestinal tract. CA125 is a surface membrane glycoprotein that is associated with ovarian cancer cells, which is positive in epithelial cells of most ovarian cancers[6
]. In women, different levels of ER and PR expression are present in most of the primary ovarian epithelial tumors. Application of the aforementioned antibodies in the detection of tissue antigens is helpful in differentiating whether the tumor originates in the appendix or the ovaries.
Thirty-three of the 35 patients in our study had CK20- and MUC-2-positive peritoneal lesions, but CK7, MUC-1, CA125, ER and PR were negative. The expression pattern of the appendix and the ovary was similar to that of the peritoneal lesions. In one of the remaining two cases, CK20, CK7 and MUC-2 were positive, and MUC-1, CA125, ER and PR were negative. The ovaries were not resected because there were no abnormal intraoperative findings. This patient’s appendix was removed at another hospital, and no specimen was examined. In the other case, the appendix appeared to be normal during surgery and was not resected. Peritoneal and ovarian lesions were negative for CK20, MUC-2, CK7, MUC-1, CA125, ER and PR. The above results suggest that PMP and ovarian lesions were implanted and metastatic appendiceal tumors in 34 of the 35 cases.
It has been reported in the literature that the appendix can be normal in patients with PMP and ovarian mucinous tumors[11
]. Lee et al[11
] have studied 196 patients with borderline ovarian mucinous tumors; of whom, only 11 had PMP. These investigators stated that the apparent absence of appendiceal lesions could have been explained by a variety of circumstances, but that the appendix was not truly normal. First, the appendix may have been left unresected because the surgeon might have observed a normal-appearing appendix during surgery. Second, even if the appendix was resected, the sample collection might not have been sufficient and complete. Finally, even if a sufficient sample was collected, tiny foci of ruptured wall might have been missed due to failure to cut serial sections. In cases with apparent absence of appendiceal lesions, we believe that lesions may have been missed.
In our study, concurrent appendiceal and ovarian lesions occurred in 20 patients, of which, 16 underwent initial surgery for appendiceal lesions, and four for ovarian tumors. Abdominal recurrence occurred at 1-85 mo after surgery, and lesions of appendiceal mucinous tumors were found in all four patients. For one patient who underwent surgery that involved the right ovary, the surgeon explored the appendix during surgery, but the appendix was not resected because no obvious abnormalities were observed by the naked eye. Mucinous tumors were found throughout the abdominal cavity in that patient at 39 mo after surgery, and the resected appendix was confirmed to contain mucinous cystadenoma.
PMP caused by pancreatic mucinous tumor occasionally has been reported in the literature[15
]. In our study, bilateral ovarian mucinous tumors were seen in one patient, accompanied by a space-occupying lesion of the gallbladder. The gallbladder was not resected due to the difficulty of the surgery. Immunohistochemical studies of the peritoneal and ovarian lesions showed that CK20, MUC-2, CK7, MUC-1, CA125, ER and PR were negative, which suggested that the ovarian tumor might be metastatic. In the literature, mucinous tumors of the gallbladder have not been reported to cause PMP, and this needs further study.
Treatment and prognosis of PMP
Treatment for PMP is complete resection of the tumors, supplemented by intraperitoneal and systemic chemotherapy. The disease often relapses, and recurrence occurs in 60%-76% of patients after surgery. Multiple surgical resections are often required, and sometimes > 20 operations are performed. Extent and invasiveness of PMP are the important causes of post-surgical recurrence. Although this disease may progress slowly, it is often fatal. The reported median survival was 5.9 years, and the 3-, 5- and 10-year survival was 81.1-83%, 50.0-81%, and 18.2-32%, respectively. Common causes of death are systemic infection secondary to wound infection, bowel obstruction, hernia, and pleural pseudomyxoma caused by tumor passing through the diaphragm. Twenty-five of the 35 patients in our study underwent two or more operations, and 11 patients died. The 5- and 10-year survival was 23.8% and 13.6%, respectively, which was lower than the survival reported in the literature. Therefore, we suggest close follow-up of patients with a diagnosis of PMP; especially those patients whose appendix has not been resected or explored during the initial surgery.
In summary, we believe that PMP often originates in the appendix, and that mucinous ovarian lesions are implanted or metastatic appendiceal tumors. Therefore, appendectomy should be performed routinely for patients in whom PMP is considered during laparotomy. The pathologist should carefully examine the gross specimen of the appendix, and collect as many tissue blocks as possible. Serial sections should be made for suspicious tissue blocks in order to search for small lesions. Additionally, because the incidence of ovarian involvement with implanted tumor is high in women with PMP, adnexa should be explored bilaterally during surgery; especially the right-side adnexa. Patients with PMP should be followed up closely; especially those whose appendix has not been resected or explored during initial surgery.