Quality of life after cytoreductive surgery and hyperthermic intra-peritoneal chemotherapy for peritoneal carcinomatosis: A systematic review and meta-analysis

Surgical Oncology 23 (2014) 199e210

Contents lists available at ScienceDirect

Surgical Oncology journal homepage: www.elsevier.com/locate/suronc

Review

Quality of life after cytoreductive surgery and hyperthermic intraperitoneal chemotherapy for peritoneal carcinomatosis: A systematic review and meta-analysis Leonard L. Shan a, Akshat Saxena c, *, Bernard L. Shan b, David L. Morris c a b c

Faculty of Medicine, Dentistry and Health Sciences, The University of Melbourne, Victoria, Australia Faculty of Medicine, Nursing and Health Sciences, Monash University, Melbourne, Australia UNSW Department of Surgery, St George Hospital, Gray Street, Kogarah, New South Wales, Australia

a r t i c l e i n f o

a b s t r a c t

Article history: Accepted 17 October 2014

Objective: To review the effect of cytoreductive surgery (CRS) and hyperthermic intra-peritoneal chemotherapy (HIPEC) on health-related quality of life (HRQOL) in patients with peritoneal carcinomatosis. Background: CRS and HIPEC is increasingly performed with curative intent for peritoneal carcinomatosis. Significant morbidity rates are reported in the context of limited life-expectancy, necessitating accurate post-operative HRQOL outcome data. Methods: A systematic review of clinical studies published after January 2000 was performed using strict eligibility criteria. Key outcomes measures were post-operative HRQOL compared to pre-operative levels and reference populations. Quality appraisal and data tabulation were performed using pre-determined forms. Data were synthesised by narrative review and random-effects meta-analysis. Tau2 and I2 values and Funnel plots were analysed for consistency and bias. Results: 15 studies (1583 patients) were included. HRQOL declines at the 3e4 month time-point before becoming similar or better compared to pre-operative levels at 1 year. The pooled-effects of combined post-operative functional assessment of cancer therapy and European organisation for research and treatment quality of life questionnaire scores were significantly improved from baseline on overall health status (p ¼ 0.001) and emotional health (p ¼ 0.001). Physical health (p ¼ 0.83), social health (p ¼ 0.48) and functional health (p ¼ 0.24) remain similar. HRQOL after 1 year is less clear, but benefits may persist up to 5 years especially on overall and physical health domains. Evidence is conflicted and inconclusive on HRQOL compared to reference populations. Levels of consistency and bias were acceptable. Conclusions: CRS and HIPEC for peritoneal carcinomatosis can confer small to medium benefits for HRQOL. These results should be interpreted with in caution due to the small studies and absence of more randomised controlled trials. © 2014 Elsevier Ltd. All rights reserved.

Keywords: Cytoreductive Intraperitoneal Hyperthermic Quality of life Survival Peritoneal carcinomatosis

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200 Rationale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200 Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200 Definition and measurement of health-related quality of life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200 Eligibility criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201 Information sources and search strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201 Study selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201

* Corresponding author. UNSW Department of Surgery, St George Hospital, Gray Street, Kogarah, New South Wales 2217, Australia. Tel.: þ61 433890393; fax: þ61 2 9113 3997. E-mail address: [email protected] (A. Saxena). http://dx.doi.org/10.1016/j.suronc.2014.10.002 0960-7404/© 2014 Elsevier Ltd. All rights reserved.

200

L.L. Shan et al. / Surgical Oncology 23 (2014) 199e210

Data items and extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202 Synthesis of results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202 Risk of bias . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204 Study characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204 Qualitative results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204 Overall health . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204 Physical health . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204 Emotional health . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205 Social health . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206 Functional health . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206 Factors affecting HRQOL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206 Non-english article . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206 Quantitative synthesis of results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206 Summary of evidence and interpretation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206 Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209 Implications for future research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209 Disclosures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209 Conflict of interest statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209 Authorship statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209

Mini-abstract Cytoreductive surgery (CRS) and hyperthermic intraperitoneal chemotherapy (HIPEC) is increasingly performed with curative intent for patients with peritoneal carcinomatosis. Significant morbidity rates are reported in the context of limited life-expectancy. This article reviews the effect of CRS and HIPEC on health-related quality of life in patients with peritoneal carcinomatosis.

Introduction Rationale Peritoneal carcinomatosis is a clinical entity whereby tumour arising from the peritoneal surface or visceral organs disseminates extensively throughout the inside surface of the abdomen [1]. It is traditionally considered a terminal condition with most patients dying at 6 months after disease onset [1,2]. Furthermore, patients' health-related quality of life (HRQOL) is greatly diminished during this time due to symptoms, and emotional and social stress [3,4]. The advent of cytoreductive surgery (CRS) and hyperthermic intraperitoneal chemotherapy (HIPEC) has changed treatment perspective towards a curative approach for selected patients. In particular, peritoneal carcinomatosis of colorectal origin and pseudomyxoma peritonei are so responsive that CRS and HIPEC are now regarded as the treatment of choice for these conditions [5,6]. Peritoneal carcinomatosis is increasingly considered to be a locoregional manifestation of disease rather than an end-stage terminal event [7]. Even though CRS and HIPEC can give patients hope of cure, it is a major operation which is known to cause substantial morbidity. Post-operative outcomes vary greatly between institutions. Perioperative mortality and morbidity in high volume tertiary centres has been reported at 0.9e5.8% and 12e52% respectively [1]. However, when taking into account all institutions, mortality and

morbidity reaches 12% and 66% respectively whilst 5-year survival rate is 15e84% [5]. Purely prolonging life may not be the best outcome for the patient [8]. HRQOL is an important measure of post-operative outcome after any surgical intervention. The importance of HRQOL assessment in CRS and HIPEC patients has been previously described [9]. Since the introduction of HRQOL assessment, more effective tools have been developed to facilitate more accurate HRQOL measurements [10]. Knowledge of HRQOL outcomes are critical in informing patient expectations and pre-operative decision making [11]. A recent narrative review reports that HRQOL after CRS and HIPEC is reported to decrease up to 6 months before returning to baseline at a year and sometimes even improving [12,13]. However, to our knowledge there has been no systematic review or metaanalysis on HRQOL after surgery. This is critical in outcome evaluation and allows both patient and clinician to assess whether the procedure will be worthwhile. Objectives We conducted a systematic review and meta-analysis of all original articles to investigate HRQOL after CRS and HIPEC in adults compared to respective patients' pre-operative status and reference populations using a time-dependent approach. Methods The structure of this systematic review followed previously recommended guidelines [14] and was written in accordance with the PRISMA checklist [15]. Definition and measurement of health-related quality of life HRQOL in peritoneal carcinomatosis is previously described and encapsulates an individual's overall, physical, emotional, social and functional health [9,16,17]. Since it is not a tangible entity, a standardised method of measurement is required which is reliable, valid, responsive, sensitive, and covers all health domains [17].

L.L. Shan et al. / Surgical Oncology 23 (2014) 199e210

201

Table 1 Outline of HRQOL scoring systems. System

Components

FACT [18]

4 main subscales. Higher scores indicate better outcome Physical well-being; Emotional well-being; Social well-being; Functional well-being; FACT-G Combination of the above subscales FACT-C FACT-G þ 9-item colon subscale TOI Physical well-being þ function well-being þ colon subscale 30 items assessing derived scores for 6 health domains and 9 symptoms; higher scores indicate better outcome Global health status; Physical function; Role function; Emotional function; Cognitive function; Social function; [below: higher scores indicate worse outcome] Fatigue; Insomnia; Appetite loss; Nausea/vomiting; Diarrhoea; Constipation; Pain; Dyspnoea; Financial problems; 38 items assessing derived scores for 2 health domains and 10 symptoms; higher scores indicate better outcome Body image; Sexual functioning [below: higher scores indicate worse outcome] Sexual enjoyment; Male sexual problems; Female sexual problems; Micturition problems; Gastrointestinal problems; Defecation problems; Stoma-related problems; Chemotherapy side effects; Weight loss; Future perspective; 0 Normal 1 Ambulatory with some symptoms 2 Requires bed rest <50% daytime hours 3 Requires bed rest >50% daytime hours 4 Completely bedridden 36 items measuring eight conceptual domains or dimensions of health. Higher scores indicate better outcome. General health (GH) Measurement of perceived overall health, including past and present health Physical functioning (PF) Indicates level of limitations in lifting, bending, kneeling, or walking moderate distance Bodily pain (BP) Represents the intensity, frequency, and duration of bodily pain and limitations in normal activities due to pain Mental health (MH) Measures the emotional, cognitive, and intellectual status of the patient Role physical (RP) Measures the degree in performing of usual activities for age and social status Role emotional (RE) Measures personal feeling of job performance at work or other activities Vitality (VT) Measures feeling of energy, fatigue, and tiredness Social functioning (SF) Indicates ability to develop and maintain mature social relationships$ Note: SF-36 can provide two summary measures e Physical Component Score (PCS) and Mental Component Score (MCS) 10-item components of the physical function domain of SF-36 Vigorous activity; Moderate activity; Lifting and carrying; Bending and kneeling; Walking several flights of stairs; Walking one flight of stairs; Walking greater than a mile; Walking several blocks; Walking one block; Bathing and dressing; 10-item rating scale of pain's effects on daily function From 0 (no pain) to 10 (as bad as you can imagine) 20-items assessing depressive symptoms in the last week Unusually bothered; appetite; can't shake off the blues; feeling of as good as other people; concentration; depressed feelings; everything feels like an effort; hopeful about future; life feels like failure; fear; insomnia; happiness; talked less than usual; lonely; feeling people are unfriendly; enjoyed life; crying spells; feeling sad; feel people's dislike; could not get going Horizontal line with markers ranging from 5 (worse) to 0 (same as before) to þ5 (better)

EORTC QLQ-C30 [19]

EORTC QLQ-C38 [20]

ECOG [42]

SF-36 [24]

ADL [22,23]

BPI [43,44] CES-D [13,25]

LAS [3]

Disease-specific HRQOL measures aim to accurately reflect a patient's experience of a specific illness or treatment. Common scoring systems for CRS and HIPEC patients are functional assessment of cancer therapy-colon specific (FACT-C) and functional assessment of cancer therapy (FACT-G) [18], European organisation for research and treatment quality of life questionnaire-cancer specific (EORTC QLQ-C30) [19], and European organisation for research and treatment quality of life questionnaire-colorectal cancer specific (EORTC QLQ-CR38) [20]. Generic HRQOL instruments are required to facilitate holistic assessment of physical, psychological, social, and functional wellbeing [21]. Common HRQOL instruments used to assess CRS and HIPEC are activities of daily living (ADL) which is derived from SF36 physical function sub-scale [22,23], medical outcomes survey short form 36 questions (SF-36) [24], and centre for epidemiologic studies-depression scale (CES-D) [13,25]. Detailed descriptions of all scoring systems and HRQOL instruments are outlined in Table 1.

carcinomatosis, (ii) disease-specific and/or generic HRQOL data recorded, and (iii) HRQOL comparisons to pre-operative status and reference populations. These studies were restricted according to the following report characteristics: (i) published after January 2000, (ii) English language, and (iii) original research only. The search period was restricted because CRS and HIPEC for peritoneal carcinomatosis is a relatively new treatment modality. Information sources and search strategy On March 2013 a literature search was conducted using MeSH keyword search on PubMed (MEDLINE) for all studies which matched the eligibility criteria above. An additional manual search of OVID (MEDLINE) and EBSCOhost (EMBASE) as well as reference lists of each included study was conducted to identify studies not covered by the initial MeSH Keyword search. All identified articles were retrieved from the aforementioned databases. Study selection

Eligibility criteria Studies considered for review had the following characteristics: (i) CRS and HIPEC for primary or secondary peritoneal

Following the search, two reviewers independently performed screening of titles and abstracts after MeSH keyword and manual searches. Studies were excluded if they did not meet eligibility

202

L.L. Shan et al. / Surgical Oncology 23 (2014) 199e210

criteria. Consensus for studies included for review was achieved by discussion between reviewers based on the pre-determined eligibility criteria. Reviewers were not blinded to any study characteristics including journal, authors and study institution. Studies were included for meta-analysis if mean ± standard deviation values for all relevant comparison groups (pre-operative, post-operative and/or reference population HRQOL) were available. Due to the small number of studies, all eligible studies were included for meta-analysis regardless of study quality. Sub-group analyses for specific health domains were performed by pooling HRQOL results of similar health domains from each instrument. Meta-analysis was not performed if fewer than 5 studies could be included. Data items and extraction All data items were pre-determined (Tables 2 and 3). Data extraction was then performed in two phases by two reviewers

using standardised pilot forms. Phase 1 assessed study quality (Table 2) and phase 2 collected results of the studies reviewed (Table 3). Synthesis of results Qualitative analysis was performed according to previous guidelines where HRQOL outcomes were categorised into physical, emotional, social and functional health domains which could be disease-specific or generic [9]. A random-effects model was used to estimate standardised mean differences for continuous data across studies. HRQOL data closest to 1 year follow-up were used in quantitative analysis as this is previously shown to be the optimum and most accurate time point for HRQOL assessment [26]. The standardised mean differences and pooled-effects (estimated overall effect [95% confidence interval]) were used as summary measures. These are depicted on

Table 2 Quality appraisal. Author year

Location

Study Patients Age design

Male Tumour origin

Disease-specific measures

CRC 25%, PMP 2%, App 23%, Gas FACT-C, FACT-G, ECOG 17%, Ov 6%, Meso 9%, Sarc 5%, SI 2% CRC 29%, App 59%, Ov 6%, Perit FACT-C, TOI, ECOG 6%

Generic HRQOL Follow-up and instrument assessment methods

Follow-up consistency

SF-36

Baseline R Follow-up: mail and phone Baseline R Follow-up: mail and phone Baseline NR Follow-up: mail

100% PR 48% RR

Baseline R Follow-up: NR Baseline R Follow-up: NR Baseline R Follow-up: clinical exam Baseline R Follow-up: mail and phone Baseline R Follow-up: clinical exam Baseline R Follow-up: mail and phone Baseline R Follow-up: NR Baseline Follow-up: mail and phone Baseline R Follow-up: mail Baseline R Follow-up: mail Baseline NR Follow-up: mail and phone Baseline R Follow-up: mail and phone

100% PR 100% RR 57% PR 100% RR 32% PR 33% RR

McQuellon [34] USA 2001

P

64

52.2 52%

McQuellon [3] 2003

USA

P

109

58.8 65%

Schmidt [36] 2005

Germany P

67

Knutsen [31] 2006 Tuttle [38] 2006 Lim [32] 2007

USA

P

5

USA

P

35

52

USA

P

28

48.8 54%

McQuellon [10] USA 2007

P

96

52.9 51%

CRC 25%, App 38%, Gas 4%, Ov 5%, Meso 9%

FACT-C, FACT-G, TOI, ECOG

CES-D, SF-36

Denmark P

23

53

35%

PMP 100%

EORTC QLQ-C30, EORTC QLQ-CR38

SF-36

Jess [30] 2008

52

40%

58.4 40% 51%

SF-36 CES-D

EORTC QLQ-C30

e

FACT-C, FACT-G, TOI CRC 20%, App 54%, Gas 6%, Meso FACT-C 9, SI 6% Sarc 100% FACT-G

e

CRC 7%, App 22%, Gas 3%, Ov 15%, Sarc 2%, SI 3% Perit 4% App 60, SI 20%

e SF-36

McQuellon [35] USA 2008

P

58

52.4 50%

App 100%

FACT-C, FACT-G, TOI, ECOG

SF-36, CES-D

Macri [33] 2009 Zenasni [26] 2009

Italy

P

17

56

CRC 41%, Gas 29%, Ov 29%

FACT-G

e

France

P

210

49.5 30%

CRC 41%, PMP 34%, App 4%, Meso 10%

EORTC QLQ-C30, EORTC QLQ-CR38

e

P

49

55

27%

PMP 100%

EORTC QLQ-C30

e

P

62

53

53%

CRC 100%

P

649

54.7 48%

Tsilimparis [37] Germany P 2013

112

56

Alves [8] UK 2010 Hill [29] USA 2011 Duckworth [11] USA 2012

24%

39%

FACT-C, FACT-G, TOI, ECOG CRC 14%, App 65%, Ov 6%, Meso FACT-C, FACT-G, 6% TOI CRC 21%, PMP 16%, App 16%, Gas 10%, Ov 19%, Meso 13%

EORTC QLQ-C30

SF-36, CES-D SF-36

e

27% PR 59% RR 37% PR 80% RR

79% PR 32% RR 91% PR 67% RR 64% PR 43% RR NR 38% PR 86% RR 100% PR 98% RR 100% PR 31% RR 35% PR 46% RR 95% PR 85% RR

Abbreviations: HRQOL e health related quality of life; NA e not applicable; NR e not recorded; OA e osteoarthritis; P e prospective; PR e participation rate; QOL e quality of life; R e retrospective; RR e response rate; ADL e activities of daily living; BPI e brief pain inventory; CES-D e centre for epidemiologic studies-depression scale; LAS e life appreciation scale; SF-36 e medical outcomes survey short form 36 questions; BP e bodily pain; GH e general health; MCS e mental component summary score; MH e mental health; PCS e physical component summary score; PF e physical functioning; RE e role emotional; RP e role physical; SF e social functioning; VT e vitality; ECOG e eastern cooperative oncology group performance status rating; EORTC QLQ-C30 e European organisation for research and treatment quality of life questionnaire-cancer specific; AL e appetite loss; C e constipation; CF e cognitive function; Di e diarrhoea; Dy e dyspnoea; EF e emotional function; F e fatigue; FP e financial problems; GHS e global health status; I e insomnia; N/V e nausea/vomiting; P e pain; PF e physical function; RF e role function: SF e social function; EORTC QLQ-CR38 e European organisation for research and treatment quality of life questionnaire-colorectal cancer specific; BI e body image; CSE e chemotherapy side effects; DP e defecation problems; FP e future perspective; FSP e female sexual problems; GP e gastrointestinal problems; MP e micturition problems; MSP e male sexual problems; SE e sexual enjoyment; SeF e sexual functioning; SRP e stoma-related problems; WL e weight loss FACT-C e functional assessment of cancer therapy-colon specific; EWB e emotional wellbeing; FWB e functional wellbeing; PWB e physical wellbeing; SFWB e social and family wellbeing; FACT-G functional assessment of cancer therapy-general score; TOI e trial outcome index.

L.L. Shan et al. / Surgical Oncology 23 (2014) 199e210

203

Table 3 Study results. Author (year)

Follow-up interval

HRQOL assessment Compared to baseline

McQuellon [34] (2001)

Disease specific Generic Function

McQuellon [3] (2003)

Disease specific Generic Function

Schmidt [36] (2005) Knutsen [31] (2006) Tuttle [38] (2006)

Lim [32] (2007)

Disease specific

Disease specific Disease specific

Disease specific Generic

McQuellon [10] (2007)

Disease specific

Generic

Function Jess [30] (2008) Disease specific Generic McQuellon [35] (2008)

Disease specific

Generic

Function Macri [33] (2009)

Disease specific

Zenasni [26] (2009)

Disease specific Generic

Alves [8] (2010)

Disease specific

Hill [29] (2011)

Disease specific

Generic

12 months

>12 months

Compared to reference population

3, 6, 12 months Yes Yes No No Improvement in physical, emotional, and functional well-being in the colon subscale and overall QOL scores. FACT-C overall QOL much worse initially, but showed improvement over 1 year. SF-36 scores showing initial decline with subsequent improvement. Initial decline in ADLs, but improved by 3 months and sustained to 12 months. 46% patients resumed more than 50% of normal activities at 6 months 5.3 ± 1.6 years No No Yes Yes Positive effect in functional well-being, physical well-being and overall QOL scores. Social and emotion function similar on FACT-C. Good to excellent general health. Majority of patient at least as good as pre-operatively. Employment: returned to work (47%), retired (24%), disabled (24%) 100% did not regret having CRS and HIPEC 4 years (1e8) No No Yes Yes Physical functioning, role functioning and social functioning improved on EORTC. Comparisons on general health, emotional functioning and cognitive functioning were not significant. 1.6 years Yes No Yes No QOL returned to baseline levels 4 months after treatment. 4, 8, 12months Yes Yes No No Except for social well-being, quality of life scores had significant or close-to-significant increases across baseline and 4, 8, and 12 months post-operatively. At 4 months after treatment, most scores had returned to pre-operative levels, and then increased at 8 and 12 months. Occurrence of adverse events correlated with significantly smaller increases in QOL scores. 2, 5, 8, 11 months Yes Yes No No Lower FACT-G scores were reported 6e8 weeks post-operatively. Despite the prolonged morbidity associated with the surgery, patients returned to baseline 3e6 months post-operatively. Physical and mental component scores of SF-36 at 3e6 months post-operatively were slightly higher than baseline. 3, 6, 12, 24 months Yes Yes Yes No Statistically significant changes over time were recorded for all subscales and the overall score, with the exception of the social well-being subscale which remained consistently high throughout the study period. FACT-G mean scores at baseline and 3 months are similar to that which has been reported as the norm for a sample from the general US adult population. Pain at its worst changes toward the better from baseline through 12 months. Nearly in every item, patients reported an increase of limitations at 3 months post-operatively, then a decrease in limitations at 12 months post-operatively. 48% participating in rigorous activity post-op. 3, 6, 12, 18, 24 Yes Yes Yes Yes months No statistically significant differences were found in the EORTC QLQ-C30 scales at any given time after surgery. Physical functioning and role physical subscales decreased at 3 months, but return to baseline at 6 months. 3, 6, 12, 24 months Yes Yes Yes No All subscales except emotional well-being returned to baseline levels or greater by 6 months. Emotional wellbeing increased from baseline to 3 months. No statistically significant differences between patients with and without disease for FACT and SF-36 scores at 1 year. Marked improvement from baseline to 12 months for physical functioning, role physical, bodily pain, and vitality subscales. The general pattern was for patients to report increased limitations at 3 months with return to baseline or better functioning at 6 and 12 months. 39% participating in rigorous activity. 3, 6 months Yes Yes No No QOL worsened at 3 months after surgery, to return at 6 months to pre-operative levels. No statistically significant variations in emotional well-being scores. Social well-being was not influenced by treatment 3, 6, 12 months No Yes No No Patients reported a good QOL. Scores referring to sexual function were particularly low. Depression following HIPEC was not significantly different from the scores of the student and cancer patient problems. 1, 3, 6, 12 months Yes Yes No No Significant improvement in pre-operative symptoms was seen in pain reported at 1 year. Fatigue, nausea or vomiting showed little improvement over the 12 months. Initial drop in global QOL 1 month post-operatively, but return to pre-operative levels at 3 months. After 12 months from surgery there was no deterioration in functional scale measures from pre-operative baseline levels. Patients who underwent major tumour debulking reported the most marked improvements in role, emotional, cognitive and social abilities. 3.2 years Yes No Yes No Significant changes were observed for emotional well-being with scores improving at 3 months and staying above baseline at 6 and 12 months. FACT-G and FACT-C total scores all dropped below baseline levels at 3 months but rebounded to above or near baseline at 6 and 12 months. Significant changes in SF-36 scores were seen for the role physical subscale. Bodily pain decreased, and social functioning decreased. Pain scores increased above baseline at 3 months, but decreased below baseline at 6 and 12 months post-operatively. (continued on next page)

204

L.L. Shan et al. / Surgical Oncology 23 (2014) 199e210

Table 3 (continued ) Author (year)

Follow-up interval

HRQOL assessment Compared to baseline

Duckworth [11] (2012)

Disease specific Generic

Tsilimparis [37] (2013) Disease specific

12 months

>12 months

Compared to reference population

1 year No Yes No Yes QOL comparable to reference FACT normative population data. SF-36 scores, physical functioning, role physical and physical component scores were significantly lower than those of the general population, while the mental health, bodily pain and mental component scores were significantly higher. 1, 6, 12, 24, 36, 48, Yes Yes Yes Yes 69 months Symptoms were an issue in the 1 month visit but mostly resolved at the 6 month measurement. Significant decrease in all elements of QOL in the early post-operative period. QOL significantly recovered at the 6 month post-operative measurement and was back at or close to baseline at 24 months post-operatively for most elements of QOL.

Abbreviations: HRQOL e health related quality of life; NA e not applicable; NR e not recorded; OA e osteoarthritis; P e prospective; PR e participation rate; QOL e quality of life; R e retrospective; RR e response rate; ADL e activities of daily living; BPI e brief pain inventory; CES-D e centre for epidemiologic studies-depression scale; LAS e life appreciation scale; SF-36 e medical outcomes survey short form 36 questions; BP e bodily pain; GH e general health; MCS e mental component summary score; MH e mental health; PCS e physical component summary score; PF e physical functioning; RE e role emotional; RP e role physical; SF e social functioning; VT e vitality; ECOG e eastern cooperative oncology group performance status rating; EORTC QLQ-C30 e European organisation for research and treatment quality of life questionnaire-cancer specific; AL e appetite loss; C e constipation; CF e cognitive function; Di e diarrhoea; Dy e dyspnoea; EF e emotional function; F e fatigue; FP e financial problems; GHS e global health status; I e insomnia; N/V e nausea/vomiting; P e pain; PF e physical function; RF e role function: SF e social function; EORTC QLQ-CR38 e European organisation for research and treatment quality of life questionnaire-colorectal cancer specific; BI e body image; CSE e chemotherapy side effects; DP e defecation problems; FP e future perspective; FSP e female sexual problems; GP e gastrointestinal problems; MP e micturition problems; MSP e male sexual problems; SE e sexual enjoyment; SeF e sexual functioning; SRP e stoma-related problems; WL e weight loss FACT-C e functional assessment of cancer therapy-colon specific; EWB e emotional wellbeing; FWB e functional wellbeing; PWB e physical wellbeing; SFWB e social and family wellbeing; FACT-G functional assessment of cancer therapy; TOI e trial outcome index; LAS e life appreciation scale; BPI e brief pain inventory; ref - reference population.

Forest plots. An estimated standard deviation could be derived from standard error and 95% confidence interval values using the calculator tool of Review Manager software. The consistency of results across studies was assessed by the Tau2 statistic for clinically relevant heterogeneity [27] and I2 statistic for statistical heterogeneity [28]. A p-value < 0.05 was considered significant for pooledeffects. All statistical analysis was performed on Review Manager (RevMan) [Windows] version 5$1 (Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2011).

A complete set of study characteristics is outlined in Tables 2 and 3 Qualitative results Complete results of qualitative analysis are provided in Table 3 with data at latest follow-up tabulated.

Study characteristics

Overall health Overall post-operative HRQOL is similar or better compared to baseline on FACT-C and FACT-G total scores at 1 year [31,32,35,38] and EORTC global health status at 1 and 2 years [8,30]. General health is similar at 2 years after surgery on SF-36 [30]. Tsilimparis et al. report a slower recovery with global health only becoming better at 3 years [37]. Overall FACT-C and general health domains on SF-36 remain improved at 5 years [3]. General health on SF-36 is worse compared to the reference population at 2 years [28,30]. In contrast, global health status on EORTC QLQ-C30 is better than reference populations at 2 and 3 years [30,37], but not at 4 years [36].

After careful systematic selection, 15 studies with a total of 1584 patients were included for review (Fig. 1) [3,8,10,11,26,29e38]. All identified articles were retrieved. All studies were designed prospectively, encompassing a variety of primary tumour origins for peritoneal carcinomatosis. Follow-up was conducted over a period of 2 monthse5.8 years. According to previous guidelines, a response rate of >85% (loss to follow-up <15%) is considered ideal for treatment received analyses [14]. This was achieved in 5 studies [8,26,31,37,38]. Since participation was voluntary, patients who did not return questionnaires were more likely to be unwilling or unable to, more ill, or deceased which may positively skew HRQOL data [35,37]. Eleven studies [8,10,29e36,38] had less than 100 patients which is an important source of bias. All studies utilised disease-specific HRQOL instruments, but only 8 had both disease-specific and generic HRQOL data [3,10,11,29,30,32,34,35]. Both methods of HRQOL assessment are important in accurately assessing the subjective nature of a patient's HRQOL.

Physical health Physical well-being subscale on FACT-C and FACT-G declines after surgery and is worse at around 3 months, but increases to be similar or better by 6 or 12 months [10,29,33e35,38]. This pattern of initial decline and subsequent improvement is supported by EORTC physical function scores [8] and SF-36 bodily pain scores [3,10,29,32,35]. These benefits for physical well-being appear to be sustained up to 5 years post-operatively [3] even though it may be 3 years before returning to baseline [37]. Jess et al. report worse bodily pain scores on SF-36 and physical function on EORTC until 24 months after surgery [30]. Patient's vitality on SF-36 declined after an initial improvement at 3 months [10,35]. Other studies show that patients' vitality on SF-36 improved slowly, but steadily after an initial decline [29,30]. Patients with intense pain on Likert scale benefited from surgery with about half as many patients experiencing the same levels of pain post-operatively [10,34]. Almost all patients report fewer difficulties on EORTC physical functioning [26]. Specific symptomatology varies between studies. Despite a

Risk of bias The risk of bias in individual studies was assessed by a qualitative review based on study quality and data tabulated in Table 2. Risk of bias across studies was also assessed by Funnel plots to assess for publication bias. Results

L.L. Shan et al. / Surgical Oncology 23 (2014) 199e210

205

Figure 1. Search algorithm and identified articles.

worsening at 1 month, pain, appetite loss, nausea, vomiting and constipation could return to baseline by 6 months [37]. Jess et al. report nausea, vomiting, sleep, appetite, diarrhoea, constipation are similar or better after surgery [30], whereas studies also show nausea, vomiting, diarrhoea, dyspnoea, fatigue and sleep disturbance were persistently worse [8,30,37]. Compared to the reference population, bodily pain and vitality was reported as better at 1 and 2 years on SF-36 [3,11,30]. EORTC QLQ-C30 physical function is similar at 2 years [30], but worse at 3 and 4 years [36,37]. Fatigue, insomnia, pain, nausea and diarrhoea were persistent and worse compared to reference populations [36]. Emotional health Post-operative emotional well-being on FACT-C is similar or better at approximately 3, 6 and 12 months [10,29,33e35,38], but is not significantly different at 5 years [3]. SF-36 role emotional and mental health domains appear to improve as a result of surgery

after an initial decline [3,10,29,32,35]. Similarly, emotional function on EORTC improved at 1 or 2 years [8,30,37]. Jess et al. report improved role emotional and mental health domains at 3 months, but worse results at 2 years [30]. Tsilimparis et al. report emotional function similar to baseline at 3 years [37]. Many patients avoid becoming clinically depressed as measured by CES-D [3,10,35] even though there may be an initial worsening of depressive symptoms at 3 months [29]. The proportion of patients with significant depressive symptoms declined from 38% to 29% at 12 months [34] and 77% patients report fewer emotional functioning difficulties after surgery [26]. Compared to the reference population, the mental health domain on SF-36 is better at 1 year post-operatively [11]. Role emotional, mental health and emotional functioning is worse on SF-36 and EORTC QLQ-C30 at 2 and 4 years respectively [30,36]. The level of depression and anxiety are not significantly different to reference population [26].

206

L.L. Shan et al. / Surgical Oncology 23 (2014) 199e210

Social health Social well-being on FACT-C, FACT-G and EORTC remains largely unchanged after surgery compared to baseline [3,10,29,33e35,37,38]. Fewer difficulties on EORTC social functioning are reported in 88% of patients [26]. Social function on SF36 may be worse at 1 year [29,35], but becomes similar to baseline at 2 years [30] and may become superior to post-operative status at 5 years [3]. Compared to the reference population, social functioning is worse or similar on SF-36 at 1 and 2 years [10,11,30] and EORTC QLQ-C30 at 1 and 4 years [8,36]. Jess et al. report slightly better social function at 2 years on EORTC [30]. At 3 years, social function appears to be much worse than reference populations [37]. Functional health Most patients reach a post-operative functional state at least as good as pre-operatively by 6 months or 12 months in functional well-being on FACT-C [10,29,34,35,38] Role function on EORTC declines at 1 month, but improves by 12 months and remains similar or better at 2 years [8,30,37]. After an initial decline in the early post-operative period, physical function [10,34,35] and role physical on SF-36 improves to be at least as good as pre-operatively [3,10,29,32,35]. This benefit may persist to 5 years [3]. However, functional well-being on Fact-G is reported to be worse at 6 months [33] and both SF-36 physical function and role physical domains may be worse at 2 years [30]. Post-operative ECOG performance status was 0 in 58e88% patients [3,10,34,35]. At 1 year, less patients were able to participate in vigorous activities and walk long distances [10,34,35]. However, more were able to climb a flight of stairs, walk short distances and bathe independently [10,34,35]. Sixty-three percent of patients report no pain with walking around at 1 year [29] and 73e85% were able to return to most of their normal activities [3,10,29,35]. Almost half resumed employment and 77% rated their energy levels as almost back to normal [3]. 76% patients reported their life was different, but better since surgery on the life satisfaction scale [3]. All patients did not regret the procedure [3]. Compared to the reference population, role function is persistently worse on EORTC QLQ-C30 and SF-36 for up to 4 years of follow-up [30,36,37]. Factors affecting HRQOL A variety of factors may affect HRQOL after CRS and HIPEC. However, only 3 studies reported these results. Thus, data is currently inconclusive [26,33,38]. Non-english article The abstract of the relevant non-english article reported similar findings those of the current review [39]. Quantitative synthesis of results A limited meta-analysis of 8 studies was performed on preoperative disease-specific HRQOL scores compared to postoperative scores at follow-up closest to 1 year [8,10,29,30,34,35,37,38]. The pooled-effects of combined post-operative FACT-C and EORTC scores were significantly improved from baseline on overall health status (p ¼ 0.001, Fig. 2A). Subgroup analysis shows improvements in emotional health (p ¼ 0.001, Fig. 2B) while physical health (p ¼ 0.83, Fig. 2C), social health (p ¼ 0.48, Fig. 2D) and functional health (p ¼ 0.24, Fig. 2E) remains similar. Tau2 (range: 0.0e0.18) indicated a low level of clinically relevant heterogeneity. Statistical heterogeneity was present on I2 (range: 0e74%) values. Overall, a good level of consistency across studies

was achieved given only 8 studies could be included for quantitative analysis. Funnel plots showed studies were closely aligned and symmetrical to the central line of pooled-effect (Fig. 3), indicating a lower likelihood of significant publication bias [28]. Discussion Summary of evidence and interpretation CRS and HIPEC has a demonstrable benefit for patients' HRQOL. Overall HRQOL returns to similar or better levels by 1e2 years compared to before surgery and these levels may be maintained for up to 5 years. Quantitative analysis demonstrated a 28% improvement in pooled-effect from baseline (Fig. 2A). This provides a simple overview of HRQOL benefits after surgery which can be easily used by clinicians during consultations. This review shows that even though patients suffered from ongoing gastrointestinal and constitutional symptoms, physical health improves. There is a noticeable decline within the first 3 months, but at some time between 6 months and 3 years, physical health becomes at least as good as pre-operatively. There was no significant difference in pooled-effect of combined FACT and EORTC scores (Fig. 2C). Given the relatively high levels of morbidity following CRS and HIPEC, the initial deterioration in physical health should be expected. Patients should be counselled about ongoing symptoms and the early decline in physical health to allow for realistic expectations of surgical outcome. A long-term perspective should be taken for physical health which is particularly important in patients who do not have at least 6 months prognosis following surgery. These results indicate further emphasis on treatment of post-operative and disease-related symptoms is required. Emotional health is the greatest beneficiary of surgery despite an initial decline likely attributed to morbidity of surgery. Patients have a similar or better emotional HRQOL as early as 3 months postoperatively. Pooled-effects of combined FACT and EORTC scores reflect this benefit with a 38% improvement (Fig. 2B). This may be the product of increased hope for prolonged survival, especially when physical recovery begins after the initial post-operative period [9]. In the context of a limited life expectancy, emotional health benefits are critical for patients and this should be stressed during pre-operative decision making. Surgery has little impact on social health domains which remain similar for up to 2 years. This is supported by Fig. 2D where there is no significant difference in combined FACT and EORTC scores. It is possible that by the time patients are being evaluated for CRS and HIPEC, the effects of diagnosis and long-term impacts on social health parameters are already fixed. Unless the patients, and more importantly their friends and families, are aware of the curative intent of surgery and post-operative results, attitudes towards the patient and their diagnosis are unlikely to change. By 1 year, functional status is at least as good as pre-operatively and this can be maintained to 5 years. Activities of daily living and satisfaction levels are high. Combined FACT and EORTC scores did not show a significant difference from baseline (Fig. 2E). Functional status is of key marker of success after surgery because it allows patients independence and the capacity to undertake their normal daily activities. Even though functional status may not necessarily be significantly improved, the fact that it remains at least similar is positive in the context of a morbid operation. Data on overall HRQOL compared to reference populations is conflicted and data is scarce. Physical health domains compared to reference populations indicate it is possible to reach a comparable level. Emotional health can exceed reference populations by 1 year, but declines in the long-term. Overall, social health is similar or worse compared to reference populations from 1 to 4 years. At 4

L.L. Shan et al. / Surgical Oncology 23 (2014) 199e210

207

Figure 2. Post-operative versus pre-operative combined FACT-C & EORTC scores 2A e overall health 2B e emotional health 2C e physical health 2D e social health 2E e functional health.

208

L.L. Shan et al. / Surgical Oncology 23 (2014) 199e210

Figure 3. Funnel plots of combined FACT-C & EORTC scores 3A e overall health 3B e emotional health 3C e physical health 3D e social health 3E e functional health.

years post-operatively, functional status may be worse than reference populations. These results are to be interpreted with caution because the comparison reference populations are heterogeneous and may not be appropriately matched populations for age, disease status and various other characteristics. Post-operative HRQOL is becoming an increasingly important and recognised outcome of surgical interventions. Recent literature is rapidly expanding on this issue and the same applies to CRS and HIPEC. McQuellon et al. have been influential for peritoneal carcinomatosis patients and previously described the details of HRQOL assessment in CRS and HIPEC for peritoneal carcinomatosis [9,16]. The short lifespan of peritoneal carcinomatosis patients elevates HRQOL to the highest priority. McQuellon et al. outline three major benefits of measuring HRQOL after surgery: (i) providing patient and clinician additional information on post-operative HRQOL to assist informed pre-operative decision making, (ii) improve patient care and symptom management, especially for poor reporters of symptoms or patients at risk of loss to follow-up, (iii) assisting survivorship planning [16]. Despite improvements in peri-operative morbidity, mortality, and long-term survival, CRS and HIPEC is a major intervention. Our review demonstrates patients must be willing to tolerate an early decline in HRQOL as a trade-off for longer survival and improved HRQOL. Guidelines recommend monitoring HRQOL after CRS and HIPEC as a standard of care because there is a great potential for harm following the procedure that impairs HRQOL which has to be

carefully balanced with the possibility for prolonged survival and possibly cure [9]. Hence assessment of HRQOL is important in gauging whether surgery is worthwhile. The greatest dilemma facing clinicians and patients is whether the prolonged survival is long enough for the patient to pass the initial recovery period and experience the positive HRQOL benefits. The HRQOL findings of this review must be viewed as an overall guideline and may not be applicable to all patients. Patients with peritoneal carcinomatosis are highly complex and have a variety of primary tumour origins, histopathological differences and extent of disease and prognosis. However, it appears HRQOL is more closely related to the type of procedure than the primary tumour pathology [29]. The degree to which patients have been previously treated with chemotherapy or resection varies prior to CRS and HIPEC. Furthermore, the superior expertise and operative outcomes of high volume tertiary centres compared to institutions where less procedures are performed is important [1]. Completeness of cytoreduction is the best assessor of prognosis after CRS and HIPEC, whereas peritoneal cancer index can be ascertained at the time of abdominal exploration [40]. The effect of both on HRQOL remains unclear [26,33,38]. This review found no clear predictors of HRQOL. It is worthwhile noting that HRQOL instruments were arbitrarily chosen by studies and the relevant reasoning unclear. The primary reason for this is that there is currently no consensus on which instrument is best for CRS and HIPEC patients. Hence, in order to

L.L. Shan et al. / Surgical Oncology 23 (2014) 199e210

minimise bias, this review had no inclusion or exclusion criteria based on type of HRQOL instrument. To our knowledge this is the first systematic review and metaanalysis on HRQOL outcomes following CRS and HIPEC and provides a synthesised modern reference when considering patients for surgery. The findings are closely supported by the results of previous narrative reviews [13,41]. In reality, patients with peritoneal carcinomatosis undergoing CRS and HIPEC have a limited life expectancy. Hence time-dependent HRQOL data are of great importance to patients as well as clinicians. Limitations Heterogeneous data prevented complete meta-analysis of disease-specific HRQOL instruments and precluded meta-analysis of generic HRQOL instruments and comparisons with reference populations. Key factors were statistical (no pre-operative data, data not expressed as mean ± standard deviation) and methodological (follow-up time point not reported, heterogeneous HRQOL scoring systems that could not be amalgamated) inconsistencies. However, there is good consistency across studies and the quantitative results are supported by the qualitative analysis of this review. Even though language bias may be present due to the English language eligibility criteria, we found only 1 article written in French was relevant to this review (Fig. 1). Despite this, the results in the English abstract were very similar to the findings of this review. Implications for future research Suggestions for future studies are outlined here. Prospective studies with pre-determined follow-up time points and consistent use of previously validated HRQOL instruments are required. This review demonstrates the lack of pre-operative HRQOL results and the need to express data as mean ± standard deviation. Multicentre involvement is ideal to increase patient numbers and minimise bias. In addition, future studies should compare HRQOL results to properly matched sample populations such as patients with peritoneal carcinomatosis who don't receive CRS and HIPEC. Given the HRQOL findings are derived from all peritoneal carcinomatosis patients, more investigation is required on defining the predictors of better HRQOL outcomes. Finally, further efforts are needed in identifying the appropriate HRQOL instruments to use for patients. Conclusions The most important findings of this review are that a broad range of health domains decline at the 3e4 month time-point before becoming similar or better at 1 year after surgery. Emotional health is the greatest beneficiary and is a critical component of HRQOL in cancer patients. Current data on HRQOL compared to reference populations is inconclusive, but likely to be worse. In the absence of large randomised controlled trials, a thorough evaluation of patients for CRS and HIPEC is still encouraged with clinical decision making tailored to each patient. Disclosures This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors. Conflict of interest statement There are no conflicts of interest to declare.

209

Authorship statement Guarantor of the integrity of the study: AS, LS, BS, DLM Study concepts: AS, LS, BS Study design: AS, LS Definition of intellectual content: AS, LS, DLM Literature research: AS, LS, BS Clinical studies: BS, LS Experimental studies: BS, LS Data acquisition: BS, LS, AS Data analysis: BS, LS, AS Statistical analysis: AS, LS Manuscript preparation: AS, BS, LS, DLM Manuscript editing: DLM, LS, AS Manuscript review: DLM, AS, LS Acknowledgements None to disclose. References [1] Chua TC, Yan TD, Saxena A, Morris DL. Should the treatment of peritoneal carcinomatosis by cytoreductive surgery and hyperthermic intraperitoneal chemotherapy still be regarded as a highly morbid procedure?: a systematic review of morbidity and mortality. Ann Surg 2009;249:900e7. [2] Glehen O, Mohamed F, Gilly FN. Peritoneal carcinomatosis from digestive tract cancer: new management by cytoreductive surgery and intraperitoneal chemohyperthermia. Lancet Oncol 2004;5:219e28. [3] McQuellon RP, Loggie BW, Lehman AB, Russell GB, Fleming RA, Shen P, et al. Long-term survivorship and quality of life after cytoreductive surgery plus intraperitoneal hyperthermic chemotherapy for peritoneal carcinomatosis. Ann Surg Oncol 2003;10:155e62. [4] Gray NM, Hall SJ, Browne S, Mcleod U, Lee AJ, Johnston M, et al. Modifiable and fixed factors predicting quality of life in people with colorectal cancer. Br J Cancer 2011;104:1697e703. [5] Roviellom F, Caruso S, Marrelli D, Pedrazzani C, Neri A, De Stefano A, et al. Treatment of peritoneal carcinomatosis with cytoreductive surgery and hyperthermic intraperitoneal chemotherapy: state of the art and future developments. Surg Oncol 2011;20:e38e54. [6] Brucher BL, Piso P, Verwaal V, Esquivel J, Deraco M, Yonemura Y, et al. Peritoneal carcinomatosis: cytoreductive surgery and HIPECeoverview and basics. Cancer Investig 2012;30:209e24. [7] Chua TC, Pelz JO, Morris DL. Surgery for colorectal peritoneal carcinomatosis. Scand J Gastroenterol 2012;47:277e85. [8] Alves S, Mohamed F, Yadegarfer G, Youssef H, Moran BJ. Prospective longitudinal study of quality of life following cytoreductive surgery and intraperitoneal chemotherapy for pseudomyxoma peritonei. Eur J Surg Oncol J Eur Soc Surg Oncol Br Assoc Surg Oncol 2010;36:1156e61. [9] McQuellon R, Gazzari C, Piso P, Swain D, Levine E. Quality of life and nutritional assessment in peritoneal surface malignancy (PSM): recommendations for care. J Surg Oncol 2008;98:300e5. [10] McQuellon RP, Danhauser SC, Russell GB, Shen P, Fenstermaker J, Stewart JH, et al. Monitoring health outcomes following cytoreductive surgery plus intraperitoneal hyperthermic chemotherapy for peritoneal carcinomatosis. Ann Surg Oncol 2007;14:1105e13. [11] Duckworth KE, McQuellon RP, Russell GB, Cashwell CS, Shen P, Stewart 4th SH, et al. Patient rated outcomes and survivorship following cytoreductive surgery plus hyperthermic intraperitoneal chemotherapy (CS þ HIPEC). J Surg Oncol 2012;106:376e80. [12] Senthil M. Assessment of clinical benefit and quality of life in patients undergoing cytoreductive surgery and hyperthermic intraperitoneal chemotherapy for management of peritoneal carcinomatosis. J Gastrointest Oncol 2013;4:3e4. [13] Zhu Y, Hanna N, Boutros C, Alexander Jr HR. Assessment of clinical benefit and quality of life in patients undergoing cytoreduction and Hyperthermic Intraperitoneal Chemotherapy (HIPEC) for management of peritoneal metastases. J Gastrointest Oncol 2013;4:62e71. [14] Wright RW, Brand RA, Dunn W, Spindler KP. How to write a systematic review. Clin Orthop Relat Res 2007;455:23e9. [15] Moher A, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 2009;6: e1000097. [16] McQuellon R, Duckworth KE. Health-related quality of life and cytoreductive surgery plus hyperthermic intraperitoneal chemotherapy. Curr Problems Cancer 2009;33:203e18. [17] Testa MA, Simonson DC. Assesment of quality-of-life outcomes. N Engl J Med 1996;334:835e40.

210

L.L. Shan et al. / Surgical Oncology 23 (2014) 199e210

[18] Cella DF, Tulsky DS, Gray G, Sarafian B, Linn E, Bonomi A, et al. The functional assessment of Cancer therapy scale: development and validation of the general measure. J Clin Oncol Official J Am Soc Clin Oncol 1993;11:570e9. [19] Groenvold M, Klee MC, Sprangers MA, Aaronson NK. Validation of the EORTC QLQ-C30 quality of life questionnaire through combined qualitative and quantitative assessment of patient-observer agreement. J Clin Epidemiol 1997;50:441e50. [20] Sprangers MA, te Velde A, Aaronson NK. The construction and testing of the EORTC colorectal cancer-specific quality of life questionnaire module (QLQCR38). European Organization for Research and Treatment of Cancer Study Group on Quality of Life. Eur J Cancer 1999;35:238e47. [21] Group W. Study protocol for the World Health Organisation project to develop a quality of life assessment instrument (WHOQOL). Qual Life Res 1993;2:153e9. [22] McHorney CA, Ware Jr JE, Raczek AE. The MOS 36-Item Short-Form Health Survey (SF-36): II. Psychometric and clinical tests of validity in measuring physical and mental health constructs. Med Care 1993;31:247e63. [23] McHorney CA, Ware Jr JE, Lu JF, Sherbourne CD. The MOS 36-item Short-Form Health Survey (SF-36): III. Tests of data quality, scaling assumptions, and reliability across diverse patient groups. Med Care 1994;32:40e66. [24] Ware Jr JE, Sherbourne CD. The MOS 36-item short-form health survey (SF36). I. Conceptual framework and item selection. Med Care 1992;30:473e83. [25] Hann D, Winter K, Jacobsen P. Measurement of depressive symptoms in cancer patients: evaluation of the Center for Epidemiological Studies Depression Scale (CES-D). J Psychosomatic Res 1999;46:437e43. [26] Zenasni F, Botella M, Elias D, Dauchy S, Boige V, Malka D, et al. The long-term impact of hyperthermic intraperitoneal chemotherapy on survivors treated for peritoneal carcinomatosis: a cross-sectional study. Support Care Cancer Official J Multinatl Assoc Support Care Cancer 2009;17:1255e61. [27] Rucker G, Schwarzer G, Carpenter JR, Schumacher M. Undue reliance on I(2) in assessing heterogeneity may mislead. BMC Med Res Methodol 2008;8:79. [28] Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ 2003;327:557e60. [29] Hill AR, McQuellon RP, Russell GB, Shen P, Stewart 4th JH, Levine EA. Survival and quality of life following cytoreductive surgery plus hyperthermic intraperitoneal chemotherapy for peritoneal carcinomatosis of colonic origin. Ann Surg Oncol 2011;18:3673e9. [30] Jess P, Iversen LH, Nielsen MB, Hansen F, Laurberg S, Rasmussen PC. Quality of life after cytoreductive surgery plus early intraperitoneal postoperative chemotherapy for pseudomyxoma peritonei: a prospective study. Dis Colon Rectum 2008;51:868e74. [31] Knutsen A, Sielaff TD, Greeno E, Tuttle TM. Staged laparoscopic infusion of hyperthermic intraperitoneal chemotherapy after cytoreductive surgery. J Gastrointest Surg Official J Soc Surg Alimentary Tract 2006;10:1038e43. [32] Lim SJ, Cormier JN, Feig BW, Mansfield PF, Benjamin RS, Griffin JR. Toxicity and outcomes associated with surgical cytoreduction and hyperthermic

[33]

[34]

[35]

[36]

[37]

[38]

[39]

[40] [41]

[42]

[43] [44]

intraperitoneal chemotherapy (HIPEC) for patients with sarcomatosis. Ann Surg Oncol 2007;14:2309e18. Macri A, Maugeri I, Trimarchi G, Caminiti R, Saffioti MC, Incardona S. Evaluation of quality of life of patients submitted to cytoreductive surgery and hyperthermic intraperitoneal chemotherapy for peritoneal carcinosis of gastrointestinal and ovarian origin and identification of factors influencing outcome. In Vivo 2009;23:147e50. McQuellon RP, Loggie BW, Fleming RA, Russell GB, Lehman AB, Rambo TD. Quality of life after intraperitoneal hyperthermic chemotherapy (IPHC) for peritoneal carcinomatosis. Eur J Surg Oncol J Eur Soc Surg Oncol Br Assoc Surg Oncol 2001;27:65e73. McQuellon RP, Russell GB, Shen P, Stewart 4th JH, Saunders W, Levine EA. Survival and health outcomes after cytoreductive surgery with intraperitoneal hyperthermic chemotherapy for disseminated peritoneal cancer of appendiceal origin. Ann Surg Oncol 2008;15:125e33. Schmidt U, Dahlke MH, Klempnauer J, Schlitt HJ, Piso P. Perioperative morbidity and quality of life in long-term survivors following cytoreductive surgery and hyperthermic intraperitoneal chemotherapy. Eur J Surg Oncol J Eur Soc Surg Oncol Br Assoc Surg Oncol 2005;31:53e8. Tsilimparis N, Bockelmann C, Raue W, Menenakos C, Perez S, Rau B. Quality of life in patients after cytoreductive surgery and hyperthermic intraperitoneal chemotherapy: is it worth the risk? Ann Surg Oncol 2013;20:226e32. Tuttle TM, Zhang Y, Greeno E, Knutsen A. Toxicity and quality of life after cytoreductive surgery plus hyperthermic intraperitoneal chemotherapy. Ann Surg Oncol 2006;13:1627e32. Lim C, Tordjmann D, Gornet JM, Nemeth J, Valleur P, Pocard M. Prospective study of quality of life after cytoreductive surgery and hyperthermic intraperitoneal chemotherapy using oxaliplatin for peritoneal carcinomatosis. Bull Du Cancer 2010;97:1053e60. Sugarbaker PH. Surgical responsibilities in the management of peritoneal carcinomatosis. J Surg Oncol 2010;101:713e24. Glockzin G, Schlitt HJ, Piso P. Peritoneal carcinomatosis: patients selection, perioperative complications and quality of life related to cytoreductive surgery and hyperthermic intraperitoneal chemotherapy. World J Surg Oncol 2009;7:5. Eton DT, Fairclough DL, Cella D, Yount SE, Bonomi P, Johnson DH, et al. Early change in patient-reported health during lung cancer chemotherapy predicts clinical outcomes beyond those predicted by baseline report: results from Eastern Cooperative Oncology Group Study 5592. J Clin Oncol Official J Am Soc Clin Oncol 2003;21:1536e43. Cleeland CS, Ryan KM. Pain assessment: global use of the brief pain inventory. Ann Acad Med Singapore 1994;23:129e38. Daut RL, Cleeland CS, Flanery RC. Development of the Wisconsin brief pain questionnaire to assess pain in cancer and other diseases. Pain 1983;17: 197e210.