Morbidity and Mortality of Cytoreduction with Intraperitoneal Chemotherapy: Outcomes
from the ACS NSQIP Database.
Bartlett EK, Meise C, Roses RE, Fraker DL, Kelz RR, Karakousis GC.
Ann Surg Oncol. 2013 Aug 29. [Epub ahead of print] PMID: 23990289
The original publication is available at the Annals website at
Morbidity and Mortality of Cytoreduction with Intraperitoneal Chemotherapy:
Outcomes from the ACS NSQIP database.
Edmund K. Bartlett,1 MD, Chelsey Meise,1 BA, Robert E. Roses,1 MD, Douglas L.
Fraker,1 MD, Rachel R. Kelz,1 MD, Giorgos C. Karakousis,1 MD.
1 Department of Surgery, University of Pennsylvania, Philadelphia, PA
Running Title: Morbidity and Mortality of HIPEC
All authors have read and approved the manuscript.
The authors declare no funding or conflicts of interest
Hospital of the University of Pennsylvania
3400 Spruce St., 4 Maloney, DSE
Philadelphia, PA 19147
In a national cohort of patients undergoing cytoreduction and intraperitoneal
chemotherapy, perioperative morbidity and mortality rates are found to be similar to
previously published series from individual high-volume institutions. Preoperative and
intraoperative factors associated with morbidity and mortality are identified.
Cytoreduction with intraperitoneal chemotherapy (IPC) for treatment of peritoneal
surface malignancies is increasingly utilized. However, the described morbidity and
mortality rates are based predominantly on the experience at high-volume centers. We
analyzed the American College of Surgeons National Surgical Quality Improvement
Program (NSQIP) database for a nationwide perspective on morbidity and mortality
associated with IPC.
The NSQIP database was queried for all patients undergoing IPC and cytoreduction from
2005-2011. Univariate and forward stepwise multivariate regression identified factors
associated with 30-day death and morbidity (D&M).
795 patients underwent IPC. Patients underwent a median of 7 operative procedures
(range 2-13). Median hospital stay was 9 days (range 2-79). 521 complications occurred
in 249 (31%) patients, and there were 19 (2.3%) mortalities. The most common
complications were bleeding (15.1%) and sepsis (14.6%). Univariate analysis identified
age≥60, ascites, weight loss, recent prior operation, albumin<3g/dl, bilirubin≥2mg/dl,
hematocrit≤30%, colon, spleen, small bowel, liver, kidney, diaphragm and gastric
resections, wound classification, operative time, and intraoperative transfusion
requirement as significantly associated with D&M. By multivariate analysis, age≥60,
preoperative albumin<3g/dl, gastrectomy, operative time, and intraoperative transfusion
requirement remained significantly associated with D&M. Particularly high D&M rates
were associated with preoperative albumin<3g/dl (58%), gastrectomy (62%), and
operative time>500 minutes (46%).
In this nationwide cohort, the D&M rate associated with cytoreduction and IPC is
consistent with other large series. Age≥60, albumin<3g/dl, gastrectomy, operative time,
and intraoperative transfusion requirement were associated with 30-day D&M. These
factors may help guide patient selection, counseling, and preoperative optimization prior
Peritoneal carcinomatosis can arise from a variety of malignancies including
gynecologic, gastrointestinal tract, and primary peritoneal cancers. In the late 1980s,
aggressive surgical cytoreduction combined with intraperitoneal chemotherapy was
popularized for treatment of those patients with disease confined to the peritoneal cavity.1
Since that time, a number of variations on the technique have been introduced in an
attempt to minimize the morbidity and maximize the therapeutic effect. Currently the
most widely used approach combines aggressive cytoreductive surgery with
intraoperative hyperthermic intraperitoneal chemotherapy (HIPEC). This modality has
now been evaluated for patients with pseudomyxoma peritoneii and appendiceal
carcinomatosis in numerous large retrospective series,2-7 as well as randomized trials in
gastric cancer8 and colorectal cancer.9-11 These studies have found cytoreduction and
HIPEC to be associated with improved outcomes in select patients. Despite this,
acceptance of the procedure has been limited, partially by the perceived high rates of
morbidity that accompany the operation.
The rates of perioperative morbidity and mortality range widely in the literature.
In patients undergoing cytoreduction and intraperitoneal chemotherapy for peritoneal
carcinomatosis of various primary origins, morbidity rates are reported from 0-52%. In a
2009 meta-analysis the overall reported rate was 29%.12 Reported mortality rates range
from 0-17% with a mean of 2.9%.12 These morbidity and mortality rates in the meta-
analysis were remarkably similar to the recent, and largest, multicenter study in patients
with pseudomyxoma peritonei which reported a perioperative morbidity rate of 24% and
mortality rate of 2%.2
Prior studies have identified number of prior operations,2 extent of
carcinomatosis,2, 3 number of intraoperative procedures,3, 13 number of anastomoses,13, 14
operative time,3, 13 extent of cytoreduction,4, 14 and chemotherapy dose4 as significantly
associated with increased risk of morbidity. Additionally, a significant learning curve has
been associated with cytoreduction and intraperitoneal chemotherapy, with one study
suggesting approximately 140 cases required before outcomes are optimized.15
Importantly, the learning curve is not just associated with the surgical technique but with
patient selection as well.6
Given the specialized nature of cytoreduction and intraperitoneal chemotherapy,
the vast majority of literature on the topic comes from a select few high volume centers.
Thus, we aimed to describe the morbidity and mortality of cytoreduction with
intraperitoneal chemotherapy in a national cohort of patients. Additionally, we looked to
define factors associated with 30-day death and morbidity that may help to guide patient
selection or preparation for the procedure. While numerous studies have defined the
procedure and disease-specific factors associated with morbidity, few have defined
patient preoperative factors well. As this procedure expands into lower volume centers, a
full understanding of the risk factors associated with the procedure may help to better
guide patient selection and potentially shorten the relatively long learning curve
traditionally associated with this procedure.
Patients were selected for inclusion in the study if they were enrolled in the
American College of Surgeons National Surgical Quality Improvement Program
Participant Use File (ACS NSQIP PUF) (FY 2005-2011) and underwent intraperitoneal
chemotherapy and a cytoreductive procedure during the same anesthetic. Intraperitoneal
chemotherapy was defined using Current Procedural Terminology (CPT) codes: 96445,
96446, or 96549, and cytoreduction was defined as any concurrent or other procedure
reflecting an intra-abdominal operation.
The ACS NSQIP PUF (FY 2005-2011) was selected to identify a national cohort
of patients undergoing intraperitoneal chemotherapy. In 2010, the ACS NSQIP database
contained 239 variables on 363,431 cases from 258 participating sites.16, 17 All
organizations participating in the ACS NSQIP are granted access to the Participant Use
File for research purposes. The ACS NSQIP database captures both inpatient and
outpatient complications for 30 days after operation. The dataset has been widely used
and is recognized as a robust clinical registry with reliable and valid data.18-20
Death and morbidity (D&M) within 30 days of cytoreduction and intraperitoneal
chemotherapy was defined as the primary outcome variable. Overall morbidity was
defined as the occurrence of at least one of the following postoperative complications:
superficial skin infection (involving only skin and subcutaneous tissue), deep surgical site
infection (involving muscle or fascia), or organ space infection (infection of areas
affected by the operation but deep to the fascia) (collectively referred to as surgical site
infection); sepsis or septic shock (sepsis complication); progressive renal failure or acute
renal failure requiring dialysis (renal complication); deep vein thrombosis or pulmonary
embolism (venous thromboembolic complication); pneumonia, intubation for greater than
48 hours, or re-intubation (respiratory complication); hemorrhage requiring transfusion of
at least 4 units of blood (bleeding complication); stroke or coma (neurologic
complication); myocardial infarction or arrest requiring resuscitation (cardiac
complication); or return to the operating room (reason for return not specified).17 Urinary
tract infection was not included as a major morbidity.
Preoperative variables included in the analysis were: sex, age (divided into <60 or
≥60 years for clinical applicability), race, functional status, body mass index (BMI;
underweight <18.5, normal = 18.5-25, overweight = 25.1-29.9, obese ≥30), hypertension,
diabetes, smoking, chronic obstructive pulmonary disease (COPD), congestive heart
failure (CHF), dialysis dependence, ascites, steroid use within 30 days, weight loss
(defined as >10% body weight in 6 months prior to surgery), chemotherapy within 30
days of surgery, radiation therapy within 90 days of surgery, and prior operation within
30 days of surgery. Laboratory values analyzed as binary variables included: creatinine
(≥2 or <2 mg/dl), albumin (≥3 or <3 g/dl), bilirubin (≥2 or <2 mg/dl), white blood cell
count (WBC≥12,000 or <12,000 cells/mcl), hematocrit (>30 or ≤30%), and international
normalized ratio (INR, ≥1.8 or <1.8).
Intraoperative variables included were: total number of procedures at the time of
cytoreduction and intraperitoneal chemotherapy, specific organ resections, operative
time, wound classification (clean, clean/contaminated, contaminated, or dirty), and
intraoperative transfusion requirement. The ACS NSQIP PUF provides information on up
to 21 procedures performed at the time of initial operation. These procedures are
described using CPT codes. In order to classify operative procedures by the organ
affected, the procedures were manually reviewed and the CPT codes were appropriately
sorted. Minor procedures such as gastrostomy tubes, jejunostomy tubes, and biopsies (ie.
liver biopsy) were excluded from the analysis.
Descriptive statistics were examined. Univariate and forward stepwise
multivariate logistic regression were performed. A p-value of less than 0.05 was
considered statistically significant. All data was transferred into STATA format using
Stat/Transfer Version 11.0 statistical program and analysis was performed using STATA
12.0/IC statistical software.21-23 This study was reviewed and deemed exempt from
approval by the University of Pennsylvania Institutional Review Board.
From 2005-2011, 795 patients underwent cytoreduction with intraperitoneal
chemotherapy in the ACS NSQIP database. The mean age of patients was 54, the slight
majority were female (54%), and the majority were white (82%). Table 1.
Patients underwent a median of 7 procedures at the time of intraperitoneal
chemotherapy. Peritoneal mass resection was the most common procedure (75% of
patients). The omentum was also frequently resected (42%). Colon resection was the
most common organ-specific procedure (49%), followed by cholecystectomy (36%),
splenectomy (28%) and small bowel resection (22%). Table 2.
The median length of hospital stay was 9 days (range 2-79). Overall, 521
complications occurred in 249 patients, for a morbidity rate of 31%. Most morbidity
events (75%) occurred within 2 weeks of surgery. Nineteen patients (2.3%) died within
30 days of surgery. All mortalities occurred within the first three weeks of surgery with
the majority occurring in week 2 or 3 (79%). Three patients died without a prior
morbidity listed. Thus 252 patients (32%) experienced a death or morbidity (D&M). The
most common complications were bleeding (15.1%), sepsis (14.6%), and surgical site
infection (11.4%). Figure 1 summarizes the postoperative morbidities.
Factors associated with increased rates of D&M and identified by univariate
analysis are listed in Table 2. Age ≥60 years (p=0.002), the presence of ascites (p=0.05),
weight loss of greater than 10% in the 6 months prior to the operation (p=0.01), and a
prior operation within 30 days (p=0.04) were identified as preoperative factors
significantly associated with higher rates of D&M. Albumin <3 g/dl (p<.0001), bilirubin
≥3 mg/dl (p=0.03), and hematocrit ≤30% (p=0.04) were preoperative laboratory values
significantly associated with D&M. Intraoperative factors associated with increased
D&M included the total number of procedures (p<.0001), prolonged operative time
(p<.0001), wound classification (p=0.003), and intraoperative transfusion requirement
(p<.0001). Additionally, individual organ resections of colon (p<.0001), spleen
(p<.0001), small bowel (p=0.006), kidney (p=0.0006), liver (p=0.006), diaphragm
(p=0.006), and stomach (p=0.0003) were associated with significantly increased D&M.
By multivariable regression, age ≥60 years remained the only preoperative patient
factor and albumin <3 g/dl the only preoperative laboratory value significantly associated
with D&M (OR=1.95 and 3.21, respectively). Gastric resection (OR=3.23), prolonged
operative time (OR=1.003), and intraoperative transfusion requirement (OR=1.01)
remained intraoperative factors significantly associated with D&M. Table 3. Patients 60
years or older had a D&M rate of 39%, and those receiving two or more units of blood
intraoperatively had a D&M rate of 38%. Particularly high D&M rates were associated
with preoperative albumin <3g/dl (58%), gastrectomy (62%), and operative time >500
minutes (46%). In the few patients 60 years or older with low preoperative albumin, the
D&M rate was 75% (12/16). Given the low incidence of mortality in the cohort, a
statistical analysis specifically for mortality was not performed. However, death more
frequently occurred in patients with preoperative albumin <3g/dl (6%, 3/52), a gastric
resection (7%, 2/27), or those who received two or more units of blood intraoperatively
(6%, 6/108). Operative time >500 minutes was not associated with particularly increased
mortality (3%, 10/290).
From 2005-2011, 795 patients in ACS NSQIP participating hospitals underwent
cytoreduction with intraperitoneal chemotherapy. The 30-day morbidity rate in these
patients was 31% and the 30-day mortality rate was 2.3%. Postoperative infection
(systemic or surgical site) and bleeding were the most frequent complications. We
identified age ≥60, albumin <3 g/dl, gastric resection, prolonged operative time, and
intraoperative transfusion requirement as significantly associated with death and
The association of increased age, prolonged operative time, and increased need
for transfusion with D&M is consistent with several previous studies which have shown
these factors to be associated with morbidity after major laparotomy.24-26 Gastric
resection as an independent factor associated with morbidity in the setting of
intraperitoneal chemotherapy, however, has not been well described. In fact, multiple
studies have looked at the morbidity associated with cytoreduction and intraperitoneal
chemotherapy in the setting of patients with gastric cancer, and a recent systematic
review found the overall morbidity to be 22% in these patients.27 Of note, many of these
patients had previously undergone resection of their primary tumor, so the rate of
gastrectomy in these studies was variable. One single institution study looked specifically
at the safety of gastric resections at the time of intraperitoneal chemotherapy across all
histologies.28 They reported a 45% rate of morbidity in 37 patients, which although high
is still less than the 62% we observe here. Given that other major organ resections such as
liver or pancreas were not associated with similar complication rates, this finding is
difficult to explain. Gastric resection is often associated with an anastomosis, and the
number of anastomoses has been identified in other series as associated with morbidity.13,
14 Additionally, underlying histology, disease burden, and patient selection may all
contribute to the high D&M rate we observed, but these variables are not reliably
captured in ACS NSQIP.
Although well described in the general surgical literature,24, 25 the finding that low
albumin is associated with markedly high rates of postoperative morbidity (58%) is
important, as it is the one potentially modifiable factor identified in this analysis. A low
preoperative albumin value may be a marker of disease status or extent of prior treatment.
Nevertheless, it should be taken into account when counseling patients prior to surgery
and in the consideration of preoperative supplemental nutrition. Given that these patients
are frequently heavily pretreated with chemotherapy in addition to the intraoperative
therapy, even the most robust patient is at risk for poor wound healing. If possible,
optimizing nutrition preoperatively may be an approach to improving outcomes in these
patients, particularly those at high risk such as the elderly and those with low albumin.
There are a number of limitations associated with the use of ACS NSQIP data for
this analysis worth discussing. Neither the histologic diagnosis nor specific pathologic
information is available for these patients (including extent of disease), yet the peritoneal
carcinomatosis index (PCI) and the extent of cytoreduction achieved have both been
associated with morbidity in prior series.2-4, 14 Despite not being directly captured in this
study, the impact of these variables is likely reflected in our study by operative time and
intraoperative transfusion requirement which were both found to be highly associated
with death and morbidity.
Additionally, details specific to the cytoreduction and intraperitoneal
chemotherapy such as the type of chemotherapy, open versus closed technique, use of
hyperthermia, and use of early postoperative chemotherapy are not captured in the
database. Finally, no data is available at the hospital or surgeon level to allow comparison
of outcomes associated with high versus low volume centers. Despite these limitations,
the identified morbidity and mortality rates are remarkably similar to the morbidity and
mortality identified in recent large series of patients with pseudomyxoma peritonei and in
a large meta-analysis of patients undergoing HIPEC.2, 12
Although studies defining the morbidity and mortality of cytoreduction with
intraperitoneal chemotherapy have been widely published, few reports analyze the
extensive preoperative patient characteristics available through ACS NSQIP.
Additionally, ACS NSQIP captures complications for 30 days, regardless of in-patient
status, which are not accurately captured in many institutional series. Further, this report
represents the first description of a national cohort of patients undergoing cytoreduction
and intraperitoneal chemotherapy. The consistency of our findings with those reported
from high volume centers suggests that the complication rates at ACS NSQIP
participating institutions as a group are comparable to rates at high volume centers,
although this may in part be influenced by the participation of a number of high volume
centers in ACS NSQIP. Regardless, the well-established reliability of the ACS NSQIP
data should reassure referring physicians that previously reported data from single
institutions is reproducible on a national level.
In conclusion, we find the overall morbidity and mortality associated with
cytoreduction and intraperitoneal chemotherapy in the ACS NSQIP database is consistent
with individual large institutions. We identify multiple factors which may aid in patient
selection and counseling, as well as one factor, low albumin, which may guide
preoperative patient optimization. By using this national cohort, the factors we identify
are institution independent. We believe this data can be used in a complimentary fashion
to the large single center experiences to aid with patient selection and preoperative
optimization in those patients undergoing cytoreduction and intraperitoneal
The American College of Surgeons National Surgical Quality Improvement Program and
the hospitals participating in the ACS NSQIP are the source of the data used herein; they
have not verified and are not responsible for the statistical validity of the data analysis or
the conclusions derived by the authors.
1. Sugarbaker PH. Surgical treatment of peritoneal carcinomatosis: 1988 Du Pont
lecture. Can J Surg 1989; 32(3):164-70.
Chua TC, Moran BJ, Sugarbaker PH, et al. Early- and long-term outcome data of
patients with pseudomyxoma peritonei from appendiceal origin treated by a
strategy of cytoreductive surgery and hyperthermic intraperitoneal chemotherapy.
J Clin Oncol 2012; 30(20):2449-56.
Glehen O, Osinsky D, Cotte E, et al. Intraperitoneal chemohyperthermia using a
closed abdominal procedure and cytoreductive surgery for the treatment of
peritoneal carcinomatosis: morbidity and mortality analysis of 216 consecutive
procedures. Ann Surg Oncol 2003; 10(8):863-9.
Kusamura S, Younan R, Baratti D, et al. Cytoreductive surgery followed by
intraperitoneal hyperthermic perfusion: analysis of morbidity and mortality in 209
peritoneal surface malignancies treated with closed abdomen technique. Cancer
Levine EA, Stewart JHt, Russell GB, et al. Cytoreductive surgery and
intraperitoneal hyperthermic chemotherapy for peritoneal surface malignancy:
experience with 501 procedures. J Am Coll Surg 2007; 204(5):943-53; discussion
Smeenk RM, Verwaal VJ, Zoetmulder FA. Learning curve of combined modality
treatment in peritoneal surface disease. Br J Surg 2007; 94(11):1408-14.
Sugarbaker PH, Alderman R, Edwards G, et al. Prospective morbidity and
mortality assessment of cytoreductive surgery plus perioperative intraperitoneal
chemotherapy to treat peritoneal dissemination of appendiceal mucinous
malignancy. Ann Surg Oncol 2006; 13(5):635-44.
Yang XJ, Huang CQ, Suo T, et al. Cytoreductive surgery and hyperthermic
intraperitoneal chemotherapy improves survival of patients with peritoneal
carcinomatosis from gastric cancer: final results of a phase III randomized clinical
trial. Ann Surg Oncol 2011; 18(6):1575-81.
Verwaal VJ, Bruin S, Boot H, et al. 8-year follow-up of randomized trial:
cytoreduction and hyperthermic intraperitoneal chemotherapy versus systemic
chemotherapy in patients with peritoneal carcinomatosis of colorectal cancer. Ann
Surg Oncol 2008; 15(9):2426-32.
Verwaal VJ, van Ruth S, de Bree E, et al. Randomized trial of cytoreduction and
hyperthermic intraperitoneal chemotherapy versus systemic chemotherapy and
palliative surgery in patients with peritoneal carcinomatosis of colorectal cancer. J
Clin Oncol 2003; 21(20):3737-43.
Elias D, Delperro JR, Sideris L, et al. Treatment of peritoneal carcinomatosis
from colorectal cancer: impact of complete cytoreductive surgery and difficulties
in conducting randomized trials. Ann Surg Oncol 2004; 11(5):518-21.
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(6):900-7.
15 Download full-text
13. Stephens AD, Alderman R, Chang D, et al. Morbidity and mortality analysis of
200 treatments with cytoreductive surgery and hyperthermic intraoperative
intraperitoneal chemotherapy using the coliseum technique. Ann Surg Oncol
Gusani NJ, Cho SW, Colovos C, et al. Aggressive surgical management of
peritoneal carcinomatosis with low mortality in a high-volume tertiary cancer
center. Ann Surg Oncol 2008; 15(3):754-63.
Kusamura S, Baratti D, Deraco M. Multidimensional analysis of the learning
curve for cytoreductive surgery and hyperthermic intraperitoneal chemotherapy in
peritoneal surface malignancies. Ann Surg 2012; 255(2):348-56.
American College of Surgeons National Surgical Quality Improvement Program.
American College of Surgeons. http://site.acsnsqip.org/, 2010.
User Guide for the 2010 Participant Use Data File. Americal College of Surgeons
National Surgical Quality Improvement Program 2011;
Daley J, Forbes MG, Young GJ, et al. Validating risk-adjusted surgical outcomes:
site visit assessment of process and structure. National VA Surgical Risk Study. J
Am Coll Surg 1997; 185(4):341-51.
Raval MV, Hamilton BH, Ingraham AM, et al. The importance of assessing both
inpatient and outpatient surgical quality. Ann Surg 2011; 253(3):611-8.
Shiloach M, Frencher SK, Jr., Steeger JE, et al. Toward robust information: data
quality and inter-rater reliability in the American College of Surgeons National
Surgical Quality Improvement Program. J Am Coll Surg 2009; 210(1):6-16.
StataCorp. Stata/IC 12.0 Statistical Software. www.stata.com. Sata Corporation
LP. College Station, TX. 1985-2011.
Circle Systems. Stat/Transfer: Version 11. Circle Systems Inc. Seattle, WA.
Juul S. F, M. An Introduction to Stata for Health Researchers. 3rd Ed. StataCorp
Aloia TA, Fahy BN, Fischer CP, et al. Predicting poor outcome following
hepatectomy: analysis of 2313 hepatectomies in the NSQIP database. HPB
(Oxford) 2009; 11(6):510-5.
Greenblatt DY, Kelly KJ, Rajamanickam V, et al. Preoperative factors predict
perioperative morbidity and mortality after pancreaticoduodenectomy. Ann Surg
Oncol 2011; 18(8):2126-35.
Grossmann EM, Longo WE, Virgo KS, et al. Morbidity and mortality of
gastrectomy for cancer in Department of Veterans Affairs Medical Centers.
Surgery 2002; 131(5):484-90.
Gill RS, Al-Adra DP, Nagendran J, et al. Treatment of gastric cancer with
peritoneal carcinomatosis by cytoreductive surgery and HIPEC: a systematic
review of survival, mortality, and morbidity. J Surg Oncol 2011; 104(6):692-8.
Piso P, Slowik P, Popp F, et al. Safety of gastric resections during cytoreductive
surgery and hyperthermic intraperitoneal chemotherapy for peritoneal
carcinomatosis. Ann Surg Oncol 2009; 16(8):2188-94.