Women with peritoneal cancer or PM from recurrent gynecologic malignancies such as ovarian cancer or fallopian tube cancer were included in the study. All patients had clinical and/or radiological evidence of PM and a previous diagnosis of ovarian cancer, fallopian tube cancer, or peritoneal cancer. Patients with extraperitoneal disease were not included in this study with the exception of isolated pleural carcinomatosis/effusion. In our institution, nutritional status is routinely recorded in all patients with PM in order to screen for malnutrition and to offer nutritional support and/or drug therapy to these patients. Clinical data were entered into a prospective registry (Ruhr-Universität Bochum, Medical Faculty Ethics Committee, approval number 15-5280). Prevalent nutritional data were obtained from all patients at baseline, i.e., at the time of their first presentation to the clinic. Follow-up nutritional data were obtained from those who underwent repeated palliative chemotherapy cycles during every cycle. Actual caloric food intake of patients was not measured. Food intake was ad libitum. Data were analyzed retrospectively by chart review.

On the day of admission, the following data were collected: 1) medical history (weight and diet changes) and gastrointestinal symptoms; 2) physical examination including body weight and estimation of body composition with RM, body fat percentage, skeletal muscle percentage, and visceral fat level by bioelectrical impedance analysis (BIA) using the Karada Scan BF 511 Body Composition Monitor according to the instructions of the manufacturer (Omron Medizintechnik, Mannheim, Germany) [9]; 3) skinfold measurement to determining the amount of subcutaneous fat using the Slim Skinfold Caliper Measure Body Fat Tester with a range from 0 to 80 mm (eaglefit GmbH, Langenau, Germany); 4) caliper body fat percentage was calculated according to the instructions of the manufacturer; 5) routine blood chemistry including C-reactive protein (CRP), total protein, albumin, and transferrin with normal ranges (CRP, <0.5 mg/dL; total protein, 6.6–8.7 g/dL; serum albumin, 3.5–5.2 g/dL; and transferrin, 200–360 mg/dL). All measurements were non-invasive except for blood samples obtained by peripheral venous puncture. Nutritional status measurements were performed by 2 study nurses.

Patients included in this study underwent a palliative intraperitoneal chemotherapy (pressurized intraperitoneal aerosol chemotherapy [PIPAC]) with cisplatin and doxorubicin as described previously [10]. Prior to intraperitoneal chemotherapy, all patients had undergone at least 2 previous lines of standard systemic chemotherapy or were unwilling or unable (as by treating physician's assessment) to undergo a further line of systemic chemotherapy. A minimum delay of 4 weeks between the last application of systemic chemotherapy and the start of intraperitoneal chemotherapy was observed. Patients were advised not to undergo systemic chemotherapy during PIPAC. Intraperitoneal chemotherapy was performed as follows. After insufflation of a 12-mmHg CO₂ pneumoperitoneum, 2 balloon safety trocars (5 and 12 mm; Applied Medical, Düsseldorf, Germany) were inserted into the abdominal wall. The peritoneal carcinomatosis index (PCI) was determined according to Jacquet and Sugarbaker [11]. Ascites volume was documented and ascites was removed. Then, a nebulizer (MIP; Capnomed, Villigendorf, Germany) was connected to an intravenous high-pressure injector (Mark7 Arterion; Medrad, Leverkusen, Germany) and inserted into the abdomen. A pressurized aerosol containing cisplatin at a dose of 7.5 mg/m² body surface in a 150 mL NaCl 0.9% solution followed by doxorubicin at a dose of 1.5 mg/m² body surface in a 50 mL NaCl 0.9% solution was applied. The pressurized chemotherapy aerosol was maintained for 30 minutes at a temperature of 37°C and scavenged via a closed line into the air waste system of the hospital. Finally, trocars were retracted and laparoscopy ended. PIPAC was repeated every 4–6 weeks until disease progression or limiting toxicity.

All p-values are 2-tailed and p-values <0.05 are considered statistically significant. Values are given as means (range) or medians (±standard deviation [SD]) where appropriate. We performed a multivariable logistic regression analysis with CAS deterioration (yes vs. no) as the dependent variable and baseline CAS (yes vs. no), baseline body mass index (BMI; ≤24 vs. >24), serum CA125 (≤1,000 vs. >1,000 U/mL), Karnofsky index (<80% vs. ≥80%), presence of ascites (yes vs. no), parenteral nutrition (yes vs. no), and tumor response (no/weak regression vs. moderate/strong/complete regression) as the independent variables. We used SPSS 22 for Windows (SPSS Inc., Chicago, IL, USA) and SigmaPlot 12.5 (Systat Software Inc., San José, CA, USA) for statistical analyses.

Eighty-four patients with peritoneal carcinomatosis from ovarian cancer (n=77), fallopian tube cancer (n=2), and peritoneal cancer (n=5) were treated between April 2014 and May 2016 and were included in this study. The mean age of the study cohort was 60.2±12.3 years. Patients had undergone a median number of 2 (range, 1–11) prior lines of systemic chemotherapy before being treated within the context of this study. Six of 84 patients were treated during the observation period with systemic chemotherapy at an outside institution. Patient characteristics of the study population are shown in Table 1. During the study period, all women underwent at least one cycle of PIPAC with cisplatin and doxorubicin. Specifically, 84, 55, 30, 17, 6, and 17 women underwent 1, 2, 3, 4, 5, and >5 cycles of chemotherapy. Thus, 84 patients were eligible for the assessment of baseline nutritional parameters and 55 patients were eligible for longitudinal analyses and for multivariable regression analysis regarding predictive factors of CAS deterioration. Fifty-two of 84 patients did not receive a complete set of at least 3 PIPACs as intended. The reasons were as follows: lost to follow-up (20/84), medical complications (18/84), and tumor progression (14/84). Two patients received additional PIPACs but these were outside the period covered by this study.

At the time of baseline evaluation, the study population was characterized by a nutritional deficit at multiple levels, a chronic inflammatory status, and a decreased energy expenditure. Nutritional data and serum levels of albumin, total protein, transferrin, and iron at baseline are shown in Table 2. Specifically, RM (1,432±172 kcal/day), visceral fat level (7.5±3.2), skeletal muscle mass (27.2%±4.6%), upper arm circumference (27.9±4.6 cm), lower leg circumference (35.1±3.9 cm), and serum parameters (albumin [3.5±0.7 g/dL; ref, 3.5–5.2], total protein [6.3±0.9 g/dL; ref, 6.6–8.6], and transferrin [202±60 mg/dL; ref, 200–360]) were below normal limits, whereas CRP (4.3±6.8 mg/dL; ref, <0.5), caliper body fat (35.7%±6.3%), and total body fat mass (35.6%±8.5%) were increased above normal limits [12131415]. Nineteen of 84 (23%) patients had CAS at baseline.

During a median time of observation of 2.4 (range, 0.3–21.7) months, deterioration of CAS as compared to stabilization/improvement of CAS was observed in 9/55 (16.4%) and 46/55 (83.6%) of patients, respectively. Parenteral nutrition support was necessary for 5/84 patients during 8 of their 20 hospital stays. Thus, the overall need for parenteral nutrition support was low (3.9%; 8/206 total stays of the 84 patients). Table 3 compares the levels of CRP, caliper body fat, arm circumference, leg circumference, and serum parameters (albumin, total protein, transferrin, and iron) during chemotherapy cycles 1 to 3, demonstrating no significant changes in the overall study population. Box plots of BMI, caliper body fat, and BIA values (i.e., RM, body fat mass, visceral fat mass, and skeletal muscle mass) during chemotherapy cycles 1 to 3 are shown in Fig. 1. Fig. 2 contains line diagrams of individual patients depicting longitudinal changes of BMI, BIA parameters body fat mass, skeletal muscle mass, RM, and visceral fat level, as well as body fat percentage determined by the caliper method over up to 11 chemotherapy cycles. This figure demonstrates an increase of RM with a concomitant decrease of the nutritional status in a minority of patients and a stabilization of these parameters in the majority of patients during palliative chemotherapy.

In a univariate analysis, baseline values of body fat mass, visceral fat level, skeletal muscle mass, caliper body fat, presence of CAS, weight, BMI, ascites, Karnofsky index, RM, CRP, parenteral nutrition support, and tumor response (all: p=not significant) were not associated with CAS deterioration. In a multivariate analysis, none of the investigated parameters was a predictor of CAS deterioration.