Is the measurement of sarcopenia associated with oncological disease in patients undergoing cytoreductive surgery and hyperthermic intraperitoneal chemotherapy?

Peritoneal malignancies are challenging cancers to manage. While cytoreductive surgery and hyperthermic intraperitoneal chemotherapy (CRS and HIPEC), may offer a cure, it is a radical procedure associated with significant morbidity. Pre‐emptive identification of deconditioned patients for optimization may mitigate surgical risk. However, the difficulty lies in identifying a cost‐effective predictive tool. Recently, there has been interest in sarcopenia, which may occur due to malignancy. The purpose of this study was to assess the utility of sarcopenia at predicting post‐operative outcomes.


Introduction
Peritoneal malignancies are challenging cancers to treat.While cytoreductive surgery with hyperthermic intra-peritoneal chemotherapy (CRS and HIPEC) may offer cure, it is also radical and morbid. 1 Major post-operative complication rates following CRS and HIPEC in contemporary literature ranges between 20% and 25% and the same studies report mortality rates of 1-8% depending on the extent of the disease and the cancer of origin. 2,3Pre-emptive identification of deconditioned, malnourished and highly symptomatic patients for pre-operative optimization may mitigate the risks associated with radical cancer surgery and this approach has gained traction in recent years. 4In order to identify patients at high risk of post-surgical morbidity, validated predictive tools are necessary.
Over the past few decades, there has been growing interest in sarcopenia 5 a term used to describe age-related decreases in muscle mass, strength and function of the body.While this may occur due to natural causes, it is now known that sarcopenia may also occur as a result of malnutrition and malignancies. 6arcopenia can be assessed in a number of ways.The most common and low-cost methods are bedside assessments of hand-grip strength, anthropometrical measures (e.g., mid-arm muscular circumference), or bio-electrical impedance analysis. 7However, more specific gold-standards is with computerized tomography (CT), where computer analysis software is used to measure the crosssectional area of psoas muscle at the third lumbar vertebrae (L3). 7ompared to other predictive tools, the attraction with sarcopenia in CRS and HIPEC patients, is that these patients generally undergo serial imaging review (CT scans) prior to surgery, making the data readily available.Additionally, as it is not uncommon for patients undergoing major cancer surgery, whether in Australia or internationally to have to travel vast distances between home and their treating hospital, a predictive tool that does not require additional face-to-face assessment is preferred and potentially cost-effective.
In patients with peritoneal malignancy, the presence of sarcopenia has been shown in the short-term to be associated with increased morbidity rates, more severe post-operative complications and increased post-surgical mortality. 3In the longer term, reduced median overall survival has also been demonstrated. 4,6Notwithstanding this, the experience and literature of sarcopenia in patients with peritoneal malignancy remains relatively sparse.Furthermore, while studies have focused on the associations between sarcopenia and surgical or oncological outcomes, there is limited understanding of the sarcopenic changes pre-and post-operatively.Improving patients' functional status and quality of life (QOL) is an important consideration and whether CRS and HIPEC reverses the sarcopenic changes that accompanies peritoneal malignancy is of interest and currently unknown.Additionally, in view of the high rates of surgical complications with CRS and HIPEC, it is also of interest to understand how sarcopenia changes or evolves with surgical complications.It may provide yet another measure to assist surgeons in optimizing patients for surgery.
Therefore, the purpose of this study was to assess the utility of sarcopenia status at predicting post-operative outcomes.The primary aims were to assess the impact of sarcopenia on oncological outcomes (Peritoneal Carcinomatosis Index (PCI), completeness of cytoreduction (CC)) and surgical outcomes (length of stay (LOS) and complications).Secondary outcomes were to determine the association between change in sarcopenia status between the preand post-operative periods and surgical outcomes.

Study design and setting
This retrospective cohort study included consecutive patients undergoing CRS and HIPEC at Royal Prince Alfred Hospital from January 2017 to June 2020.The manuscript followed the reporting recommendations from the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines. 8Ethical approval and Governance authorization were obtained from the Human Ethics Review Committee of the Sydney Local Health District (RPAH Zone Ethics Committee -Protocol No. X20-0322 and 2020/ETH01786 and SSA No. 2020/STE03155).

Participants
Patients were retrospectively identified from an electronic database with prospective collected data.All adult patients (≥18 years) undergoing CRS and HIPEC were included in the study if they had an abdominal CT scan within 3 months before surgery and between 3 and 6 months post-operatively.Patients undergoing CRS only procedures, redo CRS and HIPEC procedures, having CT scans of poor quality, or scans that could not be retrieved from the hospital's electronic imaging repository were excluded.

Measure of sarcopenia
Relevant CT scans were collected, and a radiology registrar (MH) manually extracted axial CT slices at the mid-L3 level for each patient.These were exported into a deidentified DICOM format and uploaded into Slice-o-matic™ version 4.3 (TomoVision, Canada), where the cross-sectional area (cm 2 ) of psoas skeletal muscle could be analysed.Only patients with CT scans in the hospital's repository (PACS) were accessed due to the complexity of electronically porting DICOM files into PACS.For each individual CT image, body composition components were manually 'tagged' based on known Hounsfield Units (HU)skeletal muscle (À29 to 150), visceral adipose tissue (À150 to À50), intra-muscular adipose tissue (À190 to À30), and subcutaneous adipose tissue (À190 to À30). 9 To obtain a skeletal muscle index (SMI) (cm 2 /m 2 ), total cross-sectional area of skeletal muscle was normalized by height squared.Analysis of CT images was conducted by one trained investigator (AS), with a radiology registrar (MH) for anatomical uncertainties.Calculated definitions of sarcopenia were determined from previously established sex-specific cut-offs -SMI <38.5 cm 2 /m 2 for women and < 52.4 cm 2 /m 2 for men. 9

Oncological outcomes
The primary outcomes included PCI and CC scores.The PCI is a standardized score assigned to the largest tumour nodule in 13 different sections of the abdomen, and ranges from 0 to 39. 10 A score of 0 is designated for no disease; 1 for disease measuring up to 0.5 cm; 2 for disease measuring up to 5 cm; and 3 for disease that is >5 cm or confluence of unresectable disease.The CC score is a major prognostic indicator for peritoneal metastases, and ranges from CC-0 to CC-3.Following CRS, CC-0 indicates no visible disease; CC-1 indicates tumour nodules persisting (<2.5 mm); CC-2 for tumour nodules between 2.5 mm and 2.5 cm; and CC-3 for nodules >2.5 cm or confluence of unresectable disease. 11For all pathologies (excluding PMP), CC was stratified as CC = 0 and CC>0.Due to biological and phenotypical differences between PMP and other pathologiesthe CC score was re-stratified to CC≤1 and CC>2. 12Similarly, the PCI excluding PMP was reported separate to the PCI for only PMP.

Surgical outcomes
The surgical outcomes included operation time (hours), blood loss (millilitres), post-operative complications, severity of complications, and length of stay (LOS).

Statistical methods
Descriptive analysis of demographic variables was reported separately for three groups of patients (i.e., overall, non-sarcopenia and sarcopenia).Statistical differences in oncological and surgical outcomes were analysed using Mann-Whitney U for continuous variables, and Chi-squared for categorical variables.The method of Kaplan and Meier was used to estimate survival curves.Sub-group analysis was conducted on a small subset of the population (n = 18) with imaging in the 3-to-6-month period.Included outcomes were analysis of surgical data and in-hospital surgical outcomes (LOS and complications).Statistical analysis was performed using IBM SPSS ® Statistics 23.A P-value of <0.05 was considered significant.

Patient characteristics
A total of 201 patients underwent CRS and HIPEC within the studied period.Of these, 94 (47%) patients were included.total of 107 were excluded from the study due to imaging with PET-CT scans (n = 35), CT scans of poor quality (n = 7), or not having a CT scan available from the hospital's electronic imaging repository (n = 65).Patients with PET-CT scans were excluded as CT is considered as the gold standard method for determining skeletal-muscle composition, due to its properties in differentiating fat and other soft tissues, and because it provides a means of quantifying tissue's cross-sectional area at L3. 13 A high proportion of patients (n = 65) had CT imaging in outpatient imaging centres, and their DICOM images could not easily be retrieved for analysis through Slice-o-matic™.

Sarcopenia and oncological outcomes
For patients with invasive malignancyall pathologies excluding PMP (n = 75) -14 (18.7%) patients underwent neoadjuvant chemotherapy prior to surgery, and 22 (29.3%)patients did not undergo neoadjuvant chemotherapy after multi-disciplinary team discussion.This data was unavailable for 39 (52.0%) patients (Table 2).Due to the biological and phenotypical differences of PMP, these patients (n = 19) were analysed separately as they would not routinely undergo systemic treatment.

Sub-group analysis
A small subset of the cohort (n = 18) had CT scans available 3-to-6-months after CRS and HIPEC.Of these, seven of the patients (39%) had sarcopenia pre-operatively, with five of the seven (28%) people remaining sarcopenic after surgery.For these patients with sarcopenia, 100% experienced at least one post-operative complication, while the two patients who recovered from sarcopenia reported no post-operative complication.The average time to recover from sarcopenia was 0.71 years (259 days), and time to develop sarcopenia was 0.58 years (210 days).

Discussion
In this study, rates of sarcopenia were reported at 42.6% of our total cohort (n = 94), which is comparable to previous CRS and HIPEC studies. 15A review of the literature indicates that incidence of sarcopenia in the general population ranges from 15.6% to 20.8% for healthy individuals aged <60 years, and 21.8-31.2%for those aged 60-69 years.When considering the median age of our cohort (55 years), this rate of sarcopenia is higher than that of the general population.While age-related increases in sarcopenia can occur, our results highlight the deconditioning that malignant processes can cause even if patients are not overtly symptomatic in daily living.
In the current study, other than the association between sarcopenia and overall complications between sarcopenic and nonsarcopenic patients, there were no significant differences that were identified between the cohorts in terms of operating time, blood loss, PCI score, CC scores and nature of underlying malignancy.Yet, despite having relatively comparable cohorts with comparable surgical outcomes, there is a distinct survival difference between sarcopenic and non-sarcopenic patients.The increased risk of postoperative mortality in sarcopenic patients has also been reported in patients undergoing emergency laparotomy 16 and oesophagectomies. 17It therefore highlights the utility of sarcopenia in general as a prognostication tool for surgery.
However, the more pertinent question, is whether sarcopenia can be reversed with surgery or requires targeted rehabilitation programs, thereby improving overall survival and other long-term post-operative outcomes.While not explored in this study, sarcopenia is a known risk factor for deteriorating global QOL, physical function and severe fatigue in the post-operative period. 17Due to the physical musculoskeletal decline associated with sarcopenia, patients have reduced functional ability and activities of daily living, leading to adverse impacts on QOL. 18For CRS and HIPEC patients, the radical nature of surgery and the general poor prognosis associated with peritoneal malignancy with high-grade disease necessitates early recovery of functional status and QOL to ensure optimal post-operative outcomes.
In this study, sub-group analysis was conducted for pre-and post-operative sarcopenia in a small cohort of patients (n = 18).Of interest, were two patients (11%) who recovered from sarcopenia after CRS and HIPEC.For these two patients, they experienced no post-operative complications when compared to the post-operative sarcopenia group.While no significant conclusions may be drawn from this sub-group analysis, these results may represent patients with improved biological physiology after the elimination of peritoneal disease.Conversely, two patients (11%) developed postoperative sarcopenia and experienced post-operative complications and increased LOS.While these findings support the removal of peritoneal disease to improve biological physiology, the postoperative sample size is too small to extrapolate any meaningful conclusions.Larger longitudinal studies are needed to investigate the reversibility of sarcopenia with surgery and whether this is linked to improved QOL.
In everyday practice, CT measured sarcopenia could be a readily available and feasible modality to assist in pre-operative patient  optimization.Since pre-operative CT scans are often used for standard procedure in staging and surgical planning, 19 CT measured sarcopenia could be used in addition to conventional measures of BMI, weight, or hand-grip strength.However, the major barrier to uptake of CT-measured sarcopenia is the time-consuming nature of analysis, which often requires specific computer software. 20Nevertheless, these findings advocate for an automated and standardized method to identify patients at risk or diagnosed with sarcopenia.Additionally, these findings reinforce the prognostic role that sarcopenia has on longer-term post-operative outcomes after CRS and HIPEC.Except for emergency cases or in different healthcare systems, a majority of CRS and HIPEC patients will often undergo several months of pre-operative optimization before moving into surgery.As such, CT-measured sarcopenia could assist with identifying patients who are at high risk of complications, and perhaps require targeted pre-or post-operative nutritional and musclebuilding interventions.

Strengths and limitations
A few limitations warrant acknowledgement in this study.First, the retrospective nature of this study limited the range of available patients.Many electronic health records were unavailable, and some patients had pre-operative PET-CT images, which could not be used for analysis.Second, the single-centre methodology of this study means that reproducibility of findings in other clinical areas is limited.However, these findings are considered significant in the expanding literature surrounding CRS and HIPEC.Third, not all patients who underwent a post-operative CT scan in the 3-to-6-month period were available as many of these patients are referred from regional centres where CT scans are performed remotely and not readily available.Despite these limitations, the results reinforce the multifactorial and complex process of managing CRS and HIPEC patients.Strengths of this study were the robust statistical analysis, systematic exclusion and inclusion criteria, complete follow-up data, and the relatively large number of patients by CRS and HIPEC standards.

Conclusions
Complex surgical procedures often require patient optimisation to prevent complications and improve overall survival.While patient biomarkers -BMI and weightare often used for optimisation in CRS and HIPEC, this research advocates for the addition of CT measured sarcopenia for pre-operative planning.Sarcopenia may be an indicator of advanced disease requiring further treatment, or more frequent follow-up, and is an emerging area of research that warrants further study.Larger and cross-institutional studies are required to confirm the findings of this study, and to assess the reversibility of sarcopenia after CRS and HIPEC.

Table 2
Oncological outcomes in sarcopenia and non-sarcopenia patients (n = 94) ANZ Journal of Surgery published by John Wiley & Sons Australia, Ltd on behalf of Royal Australasian College of Surgeons.