Etiological analysis of infection after CRS + HIPEC in patients with PMP

Background Cytoreductive surgery (CRS) plus hyperthermic intraperitoneal chemotherapy (HIPEC) is the standard treatment for pseudomyxoma peritonei (PMP). It can significantly prolong the survival of patients, but at the same time may increase the risk of postoperative infection. Method Patients with PMP who underwent CRS + HIPEC at our center were retrospectively analyzed. According to PMP patients, basic clinical data and relevant information of postoperative infection, we analyzed the common sites of postoperative infection, results of microbial culture and the antibiotics sensitivity. Univariate and multivariate analysis were performed to explore infection-related risk factors. Result Among the 482 patients with PMP, 82 (17.0%) patients were infected after CRS + HIPEC. The most common postoperative infection was central venous catheter (CVC) infection (8.1%), followed by abdominal-pelvic infection (5.2%). There were 29 kinds of microbes isolated from the culture (the most common was Staphylococcus epidermidis), including 13 kinds of Gram-positive bacteria, 12 kinds of Gram-negative bacteria, and 4 kinds of funguses. All the antibiotics sensitivity results showed that the most sensitive antibiotics were vancomycin to Gram-positive bacteria (98.4%), levofloxacin to Gram-negative bacteria (68.5%), and fluconazole to fungus (83.3%). Univariate and multivariate analysis revealed the infection independent risk factors as follow: intraoperative blood loss ≥ 350 mL (P = 0.019), ascites volume ≥ 300 mL (P = 0.008). Conclusion PMP patients may have increased infection risk after CRS + HIPEC, especially CVC, abdominal-pelvic and pulmonary infections. The microbial spectrum and antibiotics sensitivity results could help clinicians to take prompt prophylactic and therapeutic approaches against postoperative infection for PMP patients.


Introduction
Pseudomyxoma peritonei (PMP) is a malignant clinical syndrome characterized by the accumulation and redistribution of mucus produced by mucinous tumor cells in the peritoneal cavity [1].Most of PMP originate from mucinous tumors of the appendix, and a few originate from primary mucinous tumors of ovaries, colons and other organs [2].The incidence of PMP is approximately 2-4 cases in 1 million per year [3][4][5], and the prevalence is approximately 25.1 cases in 1 million; the male/female ratio is 1: (1.2 to 3.4) [3,4,6], the median age of onset was 62-63 years [7][8][9][10].
Nowadays, the integrated treatment which focuses on CRS + HIPEC is the main strategy for the treatment of PMP [11,12].For the selected PMP patients, standardized CRS + HIPEC can significantly improve their overall survival up to 103.4-196 months, and the 5-year and 10-year survival rates can reach 92.1% and 80.8% [13].
CRS is a complex surgical procedure, lasts a long time, and generally has a wide range of excision, which may have a great impact on the patients.The drugs of HIPEC may expose patients to the risk of immunosuppression, and splenectomy for some patients due to local invasion will also increase the risk of immunosuppression [14].Thus, PMP patients are potentially at high risk for postoperative infection.According to literature reports, the incidence of postoperative infection adverse events after CRS + HIPEC is about 21.0% ~ 43.0% [15][16][17][18][19].
This study aims to analyze the common infection sites, microbes and corresponding antibiotics sensitivity result of PMP patients after CRS + HIPEC.So as to provide reference for the treatment of patients with such postoperative infections.

Clinical data
This study was approved by the institutional review board of Beijing Shijitan Hospital, Capital Medical University (2015- [20]).All patients signed an informed consent to receive CRS + HIPEC and for the use of their clinicopathological data for further research and academic publications.
This retrospective study included 482 patients with PMP treated with CRS + HIPEC at Beijing Shijitan Hospital from May 2015 to April 2022.Data regarding the basic clinicopathological characteristics, CRS + HIPEC related information, and postoperative infection related information (results of microbial culture and antibiotic sensitivity test, etc.) were collected.

Patient selection
All patients met the criteria for CRS + HIPEC surgery [21], and the inclusion criteria were as follows: (1) Karnofsky performance status score > 60; (2) normal peripheral blood white blood cell count ≥ 3,500/mm 3 and platelet count ≥ 80,000/mm 3 ; (3) acceptable liver function, with total bilirubin ≤ 2 × the upper limit of normal (ULN) and aspartic aminotransferase and alanine aminotransferase ≤ 2 × ULN; (4) acceptable renal function, with serum creatinine ≤ 1.5 mg/dL; and (5) other major organ functions can tolerate a major operation.Major exclusion criteria include: (1) preoperative examination revealing distant metastases; (2) imaging examination indicating mesenteric contracture; and (3) the performance status and function of vital organs that cannot tolerate major surgery.

CRS + HIPEC
All CRS + HIPEC procedures were performed by the peritoneal metastasis specialist team of our center.After successful general anesthesia, a midline incision was made in the upper abdomen from the xiphoid process to the pubic symphysis to expose the abdominal cavity fully.And then, peritoneal cancer index (PCI) score was comprehensively evaluated.After CRS, the completeness of cytoreduction (CC) score was evaluated based on the residual tumor size.Open HIPEC was administered after completion of CRS, with 120 mg cisplatin + docetaxel 120 mg, or 120 mg cisplatin + mitomycin 30 mg at 43℃ for 60 min.Subsequently, functional reconstruction of digestive tract and abdominal closure were performed.

Postoperative infection
Clinical infection should be suspected in patients with postoperative symptoms such as dyspnea, painful urination, suppurative discharge in the wound or drainage tube, or fever > 38℃.Persistent increases in neutrophil counts, procalcitonin and or C-reactive protein levels 48 h after CRS + HIPEC were also considered suspected factors for infection.For patients suspected of infection, microbial culture and antibiotics sensitivity test should be carried out on any samples obtained clinically, in order to detect infection as early as possible and take targeted treatment.

CVC infection
CVC infection was defined as positive CVC tip microbial culture accompanied by chills and fever (> 38℃) or positive venous blood microbial culture (consistent with the results of CVC tip microbial culture) within 30 days after CRS + HIPEC.

Abdominal-pelvic infection
Postoperative abdominal-pelvic infection was defined as positive microbial culture of the patient's abdominal-pelvic drainage, accompanied by signs of peritonitis, fever or other infection-related symptoms within 30 days after CRS + HIPEC.

Pulmonary infection
Postoperative pulmonary infection was defined as positive microbial culture of the patient's sputum within 30 days after CRS + HIPEC, accompanied by imaging signs of infection or symptoms such as fever, cough and sputum.

Other types of infection
Postoperative surgical wound infection was defined as positive microbial culture of the patient's incision exudate within 30 days after CRS + HIPEC, accompanied by infection symptoms such as skin swelling, heat and pain around the incision.Postoperative urinary system infection was defined as positive microbial culture of midstream urine accompanied by urinary tract irritation or systemic infection symptoms such as fever and chills within 30 days after CRS + HIPEC.Postoperative positive blood culture of bacteria/fungi (infection site unknown) was defined as that the patients showed systemic infection symptoms such as fever and chills within 30 days after CRS + HIPEC with the blood culture of microbe was positive, but the infection site was still unclear after various examinations and physical examination.

Statistical analysis
Microsoft Excel 2016 and IBM SPSS Statistics for Windows, version 26.0 were used for data analysis.Measurement data were presented as median (range) or mean ± SD and analyzed by t-test or rank-sum test.Enumeration data were presented as frequencies and analyzed using the χ2 and Fisher's exact tests.Univariate and logistic regression analysis were used to analyze the independent factors influencing postoperative infection.The Kaplan-Meier method and log-rank test were used for survival analysis.Statistical significance was set at P < 0.05.

Overall survival
Overall survival (OS) was defined as the time interval from the date of clinical diagnosis to the date of death or last follow-up.

Types of infected microbe and drug susceptibility Types of infected microbe
The 82 patients who had postoperative infection were infected with a total of 29 types of bacteria and fungi, including 13 4).

Specific infectious bacteria
Among the 82 PMP patients with postoperative infection, 6 were infected with multidrug-resistant bacteria (all occurred in pulmonary infection), of which 2 were Klebsiella pneumoniae and 4 were Acinetobacter baumannii.The antibiotics sensitivity test shows that 2 cases of Klebsiella pneumoniae were only sensitive to chloramphenicol and amikacin, respectively.2 cases of Acinetobacter baumannii were all antibiotics-resistant and the other 2 cases were only sensitive to minocycline.
The factors above with P < 0.05 were incorporated into the binary Logistic regression model, and the results of multivariate analysis showed that intraoperative blood loss ≥ 350 mL (P = 0.019) and ascites volume ≥ 300 mL (P = 0.008) were independent risk factors for postoperative infection.PMP patients with intraoperative blood loss ≥ 350 mL had a 2.454 times risk of postoperative infection than those with intraoperative blood loss < 350 mL (P = 0.019, OR = 2.454, 95%CI: 1.157-5.203);For PMP patients with ascites volume ≥ 300 mL, the risk of postoperative infection was 2.192 times than those with ascites volume < 300 mL (P = 0.008, OR = 2.192, 95%CI: 1.233-3.897)(Table 7).

Discussion
In this study, the infection rate of PMP patients after CRS + HIPEC was about 17.0%.The most common infection was CVC infection (8.1%), followed by abdominal-pelvic infection (5.2%) and pulmonary infection (4.8%).Antibiotics sensitivity test revealed vancomycin as the most sensitive antibiotic for Gram-positive bacteria (98.4%), levofloxacin as most sensitive antibiotic for Gram-negative bacteria (68.5%), and fluconazole as the most sensitive antibiotic for fungi (83.3%).Univariate and multivariate analysis revealed that ascites volume ≥ 300 mL and intraoperative blood loss volume ≥ 350 mL were independent risk factors for postoperative infection.
CRS + HIPEC is the standard treatment for PMP, which can significantly improve the survival of patients with acceptable safety [11,12].However, PMP patients treated with CRS + HIPEC usually had received several operations and multicycle chemotherapy.Most of these patients have poor physical condition and are at high risk of adverse events after invasive multi-organ resection such as CRS [22].In addition, HIPEC drugs can not only kill residual tumor cells in patients' abdominal cavity, but also have drug toxicity and immunosuppressive effects, making PMP patients potentially high-risk for postoperative infection [23,24].
Previous studies have shown that the infection rate of PM patients after CRS + HIPEC was about 21.0% ~ 43.0% [15][16][17][18][19]25].The study of Arslan et al. [22] on 169 PM patients showed that the postoperative infection rate of CRS + HIPEC was 27.8%, and the most common was surgical site infection (21.3%).Smibert et al. [19] analyzed 100 patients treated with CRS + HIPEC, and the results showed that the postoperative infection rate was 43.0%, with surgical site infection being the most common (27.0%).At our center, the postoperative infection was most frequently observed in the colorectal cancer peritoneal metastases (24.3%), and least common in PMP (17.0%) (Table 8).The overall postoperative infection rate among PM patients was 20.1% which was lower than the postoperative infection rate reported in previous studies (Table 9).This could be attributed to the mature CRS + HIPEC treatment system of our center (the center has successfully completed more than 2000 CRS + HIPEC operations so far).Each PM patient will receive adequate preoperative preparation, lung function exercise, and enteral or intra intestinal nutrition support according to the nutritional status of the patient.Previous studies also identified major microbial pathogens for postoperative infection.Arslan et al. [22] analyzed 47 infected patients after CRS + HIPEC, and the microbe culture results showed that Escherichia coli (47.1%) was the most common bacteria.Valle et al. [18] studied 78 patients infected after CRS + HIPEC and found that the most common bacteria infected was Staphylococcus epidermidis (16.7%).In this study, the most common postoperative infection of PMP patients was Staphylococcus epidermidis (25.6%).Meanwhile, according to the antibiotic sensitivity test, the most common microbe infection of CVC was Staphylococcus epidermidis (16.7%), which was highly sensitive to vancomycin, linezolid, tigecycline and so on.Enterococcus faecalis (9.8%) was the most common microbe isolated from abdominal-pelvic infection, with high sensitivity to vancomycin, penicillin G, ampicillin.The most common microbe isolated from pulmonary infection was Acinetobacter baumannii (15.9%), while minocycline and sulfamethoxazole were the only relatively sensitive antibiotics.In this study, 6 cases of PMP patients infected with multi-drug resistant bacteria, all of which were pulmonary infections, suggesting that after CRS + HIPEC, it is especially necessary to pay much attention to the pulmonary function of patients, promote sputum discharge and reduce the risk of pulmonary infection.Some studies had shown that [22,26], the main cause of death due to infection in patients after CRS + HIPEC is Candida albicans infection.In this study, none of PMP patients died  or became critically ill due to Candida albicans infection, and antibiotics tests showed that Candida albicans was very sensitive to fluconazole and voriconazole.There were several studies about postoperative infection of CRS + HIPEC revealed the following 5 major risk factors, including colorectal resection, small intestine resection, intraoperative blood loss, operation duration > 10 h, and preoperative nutritional status [18,19,25].In comparison, our study only found 2 independent risk factors for postoperative infection, intraoperative blood loss ≥ 350 mL and ascites volume ≥ 300 mL.Patients with ascites tended to have abdominal distension, poor appetite and other gastrointestinal symptoms, and the nutritional status of such patients were usually poor.However, nutritional status had been shown to have a significant impact on the immune system, and patients with impaired immune response were more likely to develop postoperative complications after gastrointestinal surgery [20,27].There was a bacteria hypothesis in the mucin formation and tumor progression in PMP, Semino-Mora et al. [28].found that the overall bacterial density of appendixes in PMP patients was much higher than in healthy people.That is probably an explanation to why ascites is associated with the infection risk, because ascites could be produced partially by bacteria.CRS + HIPEC often involved partial resection of the invaded gastrointestinal tract, for which ERAS guidelines recommend preoperative mechanical gastrointestinal preparation with or without oral antibiotics to reduce postoperative infection rates [29].
Postoperative infection was the main cause of increased length of stay and perioperative mortality in patients treated with CRS + HIPEC, as well as increased treatment costs for patients [8,30].However, no studies have shown that postoperative infection of CRS + HIPEC was associated with long-term outcome of patients.The results of survival analysis in this study also showed that there had no difference on median overall survival between PMP patients with or without postoperative infection.
This study has the following limitations: First, the types and characteristics of microbes isolated from the infected patients may be different in different CRS + HIPEC treatment centers, so it is necessary to combine infectionrelated data from multiple centers in the future to make a summary.Second, this was a single-center retrospective case-control study with a moderate sample size, and higher-level studies must verify the conclusions.
In conclusion, PMP patients may have increased infection risk after CRS + HIPEC, especially CVC, abdominalpelvic and pulmonary infections.This study analyzed the common infection sites and microbes in PMP patients after CRS + HIPEC, as well as the corresponding antibiotics sensitivity test results, which may provide reference for the early clinical empirical antibiotics use in patients with such CRS + HIPEC postoperative infection.
informed consent to receive CRS + HIPEC and for the use of their clinicopathological data for further research and academic publications.

Table 3
The types and proportion of the microbes isolated from PMP patients with postoperative infection

Table 4
The common microbes isolated from the postoperative infection sites of PMP patients

Table 5
The results of antibiotics sensitivity to microbes isolated from PMP patients with postoperative infection

Table 6
The univariate analysis of PMP patients with postoperative infection PCI Peritoneal cancer index, CC Completeness of cytoreduction, RBC Red blood cells

Table 7
The multivariate analysis of PMP patients with postoperative infection Fig. 1 Survival analysis.A Overall survival analysis of all PMP patients; (B) Survival curve analysis of infected group and non-infected group

Table 8
The comparison of postoperative infection rate after CRS + HIPEC for different PM patients at our center