Peritoneal carcinomatosis
What it is, why some cases are operated on with curative intent, what the published figures say for each tumour of origin, and how it is decided whether you are a candidate.
The peritoneum is a membrane — a very thin, moist sheet that lines the walls of the abdomen from within and folds over on itself to wrap everything inside. It weighs barely half a kilo and, spread out on a table, would cover almost as much as your skin: around two square metres.
Peritoneal carcinomatosis means there are tumour cells spread across that surface. The name describes where the disease is, not what disease it is. It is also called peritoneal metastases.
For decades it was regarded as a terminal situation with nothing to offer beyond chemotherapy. That changed — not for everyone at once and not to the same degree. It changed first for some tumours of origin and has still not changed for others. This page says which is which, with the numbers and where they come from.
A cancer that has spread to the liver or the lung travelled through the bloodstream and, as a rule, stopped being operable. One that has spread to the peritoneum has done something else: it has stayed inside a room — a large room, but a room.
That difference is what opens the door to surgery. If all the disease sits on a surface that can be reached, it can be removed in full. And if it is removed in full, the prognosis changes category.
The peritoneum is built precisely not to make itself felt, and it does its job. That is why a disease living in it can be in twenty places at once and hurt in none of them.
When symptoms do come, they tend to be indirect:
Clinical presentation as described by the author in his patient book on peritoneal carcinomatosis (Madrid, 2026).
Two numbers matter more than any others in this disease, and they are worth understanding because you will hear them.
The first is the PCI, the peritoneal cancer index. The abdomen is divided into thirteen regions; in each one the largest nodule is scored from 0 to 3, and the thirteen scores are added. The result runs from 0 to 39. A 4 means little disease, well localised; a 30 means there is something almost everywhere.
One caution: the same number means different things in different diseases. A PCI of 25 in pseudomyxoma is routine and is operated on; in gastric cancer it is a contraindication. The number measures volume, not biology.
The second is the CC, the completeness of cytoreduction score, and it describes what the surgeon left behind on closing. CC-0 is no visible disease. CC-1 is nodules up to 2.5 millimetres. CC-2 and CC-3 are more than that, and do not count as a complete cytoreduction.
The cut-off is not arbitrary: intraperitoneal chemotherapy penetrates only a few millimetres into tissue, so below that thickness it can reach what remains, and above it, it cannot. Hence the sentence that summarises the whole technique — heated chemotherapy treats what cannot be seen, after surgery has removed everything that can. It does not replace it.
What that is worth in practice: in the European series of 506 patients with colorectal origin, median survival was 32.4 months when cytoreduction was complete and 8.4 months when macroscopic disease was left behind. The same disease, the same surgeons, different results according to what stayed inside.
PCI definition: Jacquet P, Sugarbaker PH. Cancer Treatment and Research, 1996. European multicentre series of 506 patients: Glehen O, et al. Journal of Clinical Oncology, 2004.
“Peritoneal carcinomatosis” is not one disease: it is the same anatomical situation produced by very different tumours, and the prognosis depends above all on which one. These are the published data.
Colorectal origin. The trial that opened the field compared cytoreductive surgery with intraperitoneal chemotherapy against systemic treatment, with medians of 22.4 versus 12.6 months. Later series, with better selection and technique, give medians around 30 months and five-year survival of 27%. In the most recent trial, in which every patient was operated on until the abdomen was clear, the median was 41.7 months.
Ovarian origin. In stage III, adding heated intraperitoneal chemotherapy to interval surgery after neoadjuvant chemotherapy improved median survival from 33.9 to 45.7 months. At ten years, survival was 16.1% versus 10.9%. With small print that matters: that result applies to interval surgery, to stage III, and only where surgery leaves the abdomen clear or nearly so.
Appendiceal origin and pseudomyxoma peritonei. This is where results are best, and by a distance. With complete cytoreduction and low-grade histology, five-year survival was 86%. In a pooled analysis of 2,298 patients treated at sixteen centres, median survival was 196 months and 63% were alive at ten years.
Peritoneal mesothelioma. With systemic treatment alone, historical median survival is six to twelve months. In the international registry of 405 operated patients, the median was 53 months and 47% were alive at five years. The factors that counted were epithelioid subtype, absence of nodal involvement, complete cytoreduction and HIPEC.
Gastric origin. Here one has to be direct, because this is where the operation is most often offered without data to support it. The trial comparing surgery against systemic treatment in limited peritoneal disease gave 15.7 versus 16.6 months, with serious adverse events in 44% of the surgical arm against 6%, and closed early for futility. Outside a structured trial, cytoreductive surgery for gastric peritoneal metastases is not defensible today.
Colorectal: Verwaal VJ, et al. J Clin Oncol 2003; Elias D, et al. J Clin Oncol 2010; Quénet F, et al. Lancet Oncol 2021 (PRODIGE 7). Ovarian: van Driel WJ, et al. N Engl J Med 2018 and Aronson SL, et al. Lancet Oncol 2023 (OVHIPEC-1). Appendiceal: Sugarbaker PH, Chang D. Ann Surg Oncol 1999; pooled analysis of 2,298 patients across 16 centres. Mesothelioma: Yan TD, et al. J Clin Oncol 2009. Gastric: Quik JSE, et al. Lancet Oncol 2026 (PERISCOPE II).
If you have searched online, you have probably run into this trial, or into headlines saying HIPEC does not work. It deserves an honest explanation, because it is the point most often told badly.
PRODIGE 7 operated on every patient until no visible disease remained. Half then received the heated bath with oxaliplatin for thirty minutes, and half were closed without it. Median survival was 41.7 months with the bath and 41.2 without: essentially the same. Serious adverse events at sixty days were more frequent with the bath, 26% against 15%.
What that result isolates is not that the operation does not work. Quite the opposite: both arms lived far longer than these patients lived before this surgery existed, and both arms had been operated on. What the trial tells us is which of the two acts does the work. Complete cytoreduction is the therapeutic act; the oxaliplatin bath, in that specific schedule, added no survival.
And what it does not say: it did not test mitomycin C, it did not test longer perfusions, and it did not test other tumours of origin. A trial is read by drug, dose, schedule and histology, or it is not read at all.
Quénet F, et al. Cytoreductive surgery plus hyperthermic intraperitoneal chemotherapy versus cytoreductive surgery alone for colorectal peritoneal metastases (PRODIGE 7). Lancet Oncology, 2021. The trial reports its confidence interval at 95.37% because of the alpha spent at the interim analysis.
The deciding question is not “how much disease is there?” but “can all of it be removed?”. If the answer is no, the operation loses its rationale, because the benefit depends on completing it.
Answering it takes three things: knowing the origin and biology of the tumour, seeing the extent and the distribution, and judging whether your general condition allows a long operation. CT helps but falls short for small-volume disease, particularly over the small bowel, so in many cases the assessment is completed with a diagnostic laparoscopy.
The small bowel is usually what decides. Extensive disease over its loops is the most common reason a complete cytoreduction cannot be achieved, because it cannot be resected without leaving the patient without enough bowel.
The decision is made in a multidisciplinary board with medical oncology, radiology and pathology. And one warning worth hearing early: if you are not offered surgery and someone else with the same word in their report is, that is not arbitrary. The word in the report describes where the disease is, not what disease it is or how much of it there is.
Limitations of CT and the role of diagnostic laparoscopy: set out in the author's monograph on peritoneal carcinomatosis, drawing on the series cited here.
It is a long operation. In published trials the means run from six to eight hours, and in practice it depends on how many resections are needed. Disease is being removed from two square metres of surface, millimetre by millimetre. That is all that is happening, and that is why it takes time.
Afterwards come one or two days in intensive care and a nasogastric tube until the bowel wakes up, which takes three to five days. Discharge is usually between day eight and day fourteen if there are no complications. Published mean hospital stays in the trials run from thirteen to eighteen days.
The complication watched most closely is a leak from a bowel anastomosis, between the fourth and the seventh day, which often means a second operation. Perioperative mortality in published series runs between 1% and 4%, and reached 8% in the pioneering 2003 trial, with the technique and selection of the time.
And after discharge, the part nobody describes: three to six months to get back to normal life. Fatigue is the dominant complaint — it is not like being short of sleep, and it lifts over weeks. You eat little and often, you lose weight, and putting it back on takes months. Walking every day is the intervention with the best effort-to-result ratio in the whole recovery. By the third or fourth month the energy comes back.
Operating time and hospital stay: values published in PRODIGE 7 (Lancet Oncol 2021) and in the multicentre series cited. Perioperative mortality: 4.0% in Glehen 2004 (506 patients), 2.7% in Sugarbaker 1999, 2% in Yan 2009, 8% in the surgical arm of Verwaal 2003. Postoperative course and recovery: the author's practice.
This is volume surgery. In PRODIGE 7, 71% of patients were recruited at three centres that had each performed more than five hundred procedures; the results you read come from places like that.
Spain designates reference centres for this disease, the CSUR network. Hospital General Universitario Gregorio Marañón is one of them for peritoneal carcinomatosis.
If you have been told there is nothing to be done, or you have been offered an operation and want it checked, a second opinion at a high-volume unit is reasonable and offends no one. What you need to bring is imaging in digital form, the pathology, and the notes from any previous surgery.
Recruitment by centre: PRODIGE 7 (Lancet Oncology, 2021).
The figures you have read are medians and percentages from groups of patients. A median of twenty-two months does not mean “you have twenty-two months”. It means half that group went past it and half did not. Where you fall on that curve is not something the number can tell you.
And this does not replace a consultation. None of the three things that decide — the tumour of origin, the distribution of disease and your general condition — can be assessed from a screen.
Frequently asked questions
Send imaging and reports ahead of the visit so the case can be reviewed properly. Patients from outside Madrid and from abroad are seen.