Ann Surg Oncol DOI 10.1245/s10434-016-5109-3

ORIGINAL ARTICLE – GASTROINTESTINAL ONCOLOGY

Factors Associated with Readmission After Cytoreductive Surgery and Hyperthermic Intraperitoneal Chemotherapy for Peritoneal Carcinomatosis Alexander S. Martin, Daniel E. Abbott, MD, Dennis Hanseman, Jonathan E. Sussman, Alexander Kenkel, Parker Greiwe, Noor Saeed, Samar H. Ahmad, Jeffrey J. Sussman, MD, and Syed A. Ahmad, MD Division of Surgical Oncology, Department of Surgery, University of Cincinnati College of Medicine, Cincinnati, OH

ABSTRACT Purpose. Cytoreductive surgery/hyperthermic intraperitoneal chemotherapy (CRS/HIPEC) for peritoneal carcinomatosis is a morbid endeavor. Despite improvement in perioperative management of these patients, there are subsets of patients requiring hospital readmission after discharge. We sought to identify variables associated with readmission rates for CRS/HIPEC. Methods. We conducted a retrospective review of CRS/ HIPEC cases at the University of Cincinnati between 1999 and 2014. Patient-, tumor-, and treatment-specific characteristics were evaluated. The association between patientand outcome-specific variables for 30- and 90-day readmission were evaluated. Results. Of 215 CRS/HIPEC patients, the 7-, 30-, and 90day readmission rates were 9.8 % (n = 21), 14.9 % (n = 32), and 21.4 % (n = 46), respectively. The most common reasons for readmission within 90 days included abdominal pain (n = 14), intra-abdominal abscess (n = 9), malnutrition/failure to thrive (n = 8), and bowel obstruction (n = 7). The primary factor associated with readmission at all time points (7, 30, and 90 days) was the presence of an enterocutaneous fistula (p \ 0.01). Six patients (2.8 %) had multiple readmissions; 3 of these had ECF. Factors not associated with higher admission rates included sex, age, race, operative blood loss, pancreatectomy or liver resection

Presented at the 2015 Society of Surgery for the Alimentary Tract Meeting.  Society of Surgical Oncology 2016 First Received: 23 October 2015 S. A. Ahmad, MD e-mail: [email protected]

at the index operation, and postoperative complications of wound infection, line infection, and thromboembolic events. Conclusions. In patients undergoing CRS/HIPEC, readmission was primarily associated with poor pain control, malnutrition, and infectious complications. Patients with enterocutaneous fistula were also disproportionately readmitted multiple times. These data should inform clinicians about patients at high risk for readmission after CRS/ HIPEC and encourage more comprehensive coordination of postdischarge planning and care for specific patient populations.

The Centers for Medicare and Medicaid have proposed a 3 % reduction in the total reimbursements for hospitals with excessive readmissions, based on the fact that these readmissions cost the U.S. health system more than $17 billion per year. Although only a select number of surgical procedures are having their readmission rates carefully monitored, expanding their scope is on the horizon. On this basis, and with the knowledge that complex surgeries have even higher readmission rates, it is crucial to determine the factors contributing to patients returning to the hospital after discharge. It is well documented that surgeries including cystectomy, esophagectomy, pulmonary resection, and pancreatectomy have a high risk of hospital readmission. Recent 30- and 90-day readmission rates for these major cancer surgeries were found to range from 10 to 25 and 23 to 43 %, respectively.1 Cytoreductive surgery/hyperthermic intraperitoneal chemotherapy (CRS/HIPEC) is well established in managing patients with peritoneal metastases and peritoneal-based malignancies. These procedures are associated with major morbidity and mortality rates of 27–56 and 0–11 %, respectively.2 Common complications after this procedure include enterocutaneous fistula (ECF), postoperative

A. S. Martin et al.

bleeding, hematologic toxicity, and infectious complications.3 However, few studies have evaluated readmission rates after this difficult operation. Therefore, we sought to review our experience with readmission rates and factors attributed to readmission after CRS/HIPEC in patients with peritoneal carcinomatosis. METHODS Database A retrospective review of patients who underwent CRS/ HIPEC at our institution was performed. Approval for the study was obtained from the institutional review board. Fellowship-trained surgical oncologists performed all of the operations. The technique for CRS/HIPEC has been described previously by our group.2 Patient medical records were reviewed to obtain patient-, tumor-, treatment-, and perioperative-specific characteristics. Readmission was classified into three categories: readmission to any hospital (University of Cincinnati Medical Center or outside hospital) within 7, 30, or 90 days after hospital discharge. Quantitative Measures of Patient Disease The peritoneal cancer index (PCI) was calculated at the time of CRS/HIPEC for each patient using standard parameters for determining PCI as previously described by Sugarbaker.4,5 However, if the PCI was not noted at the time of surgery, then a single investigator retrospectively determined the PCI through detailed examination of operative reports along with pathology reports to confirm the size and histology of submitted specimens. A completeness of cytoreduction (CC) score was obtained to define volume of residual tumor after CRS. The standardized values included: CC0 = no gross disease, CC1 = residual tumor nodules \2.5 mm, CC2 = residual tumor nodules between 2.5 and 25 mm, and CC3 = residual tumor nodules [25 mm in size.4 Statistical Analysis Statistical analysis was performed by SAS software (SAS Institute). Chi square tests were used to assess the relationships between patient and outcome-specific variables and 7-, 30-, and 90-day hospital readmission. Odds ratios were estimated for independent variables using a regression coefficient. Multivariate analysis using a logistic regression model was performed for variables showing univariate statistical significance.

RESULTS Patient and Tumor Demographics Between December 1, 1999, and April 1, 2014, a total of 302 CRS/HIPEC procedures were performed. As a result of incomplete data, 215 cases in 203 patients were included in the final analysis. Twelve of these patients underwent multiple CRS/HIPEC procedures. The entire cohort comprised 44 % men (n = 90) and 56 % women (n = 113). Fifty-four percent of patients (n = 109) were between 40 and 60 years old. Primary tumor was of appendiceal origin in the majority of patients (80 %, n = 163), with the remaining patients having colon (7 %, n = 15), ovarian (3 %, n = 7), gastric (2 %, n = 5), and mesothelial (2 %, n = 4) cancers. Nine patients had tumors of unknown origin. The majority of tumors (n = 172) were deemed to have mucinous histopathologic characteristics on pathologic review. Perioperative Characteristics The median estimated blood loss was 300 ml [interquartile range (IQR) 200–600 ml]. Median duration of surgery was 307 min (IQR 305.5–495 min), and 46 patients required perioperative blood transfusion during the surgery (21 %; mean transfusion of 3.2 units). The most common complications after CRS/HIPEC included pulmonary complications (pleural effusion, respiratory failure, pneumothorax, pneumonia; n = 22, 10.8 %), thromboembolic events (n = 14, 6.9 %), wound infection (n = 11, 5.4 %), intra-abdominal abscess (n = 11, 5.4 %), cardiac events (atrial fibrillation, supraventricular tachycardia and pulseless electrical activity arrest, total n = 9, 4.4 %), ECF (n = 7, 3.4 %), and central venous catheter infections (n = 2, 0.9 %). Of the 203 individual patients who underwent CRS/ HIPEC, the median PCI score for the entire cohort was 13 (IQR 6–18). There was no significant difference in PCI score among readmitted (mean PCI 13, range 6–18) and those patients not readmitted (mean PCI 12, range 6–17) (p [ 0.05). There was no relationship between the PCI score and the development of complications. In the entire cohort, a complete cytoreduction (CC0) occurred in 63.5 % of patients (n = 129). CC1, CC2, and CC3 occurred in 21.7 % (n = 44), 7.4 % (n = 15), and 6.4 % (n = 13), respectively. There were no significant differences in CC scores between readmitted and nonreadmitted patients (p = 0.058). A complete cytoreduction (CC0) occurred in 64.8 % (n = 103) of nonreadmitted patients and in 59.1 % (n = 26) of readmitted patients. Of the readmitted patients, 27.3 % (n = 12) had a CC1 reduction, while 20.1 % (n = 32) of nonreadmitted

Factors Associated with Readmission After Cytoreductive Surgery…

patients had a CC1 reduction. The CC2 and CC3 were 4.6 % (n = 2) and 4.6 (n = 2) for readmitted patients. For nonreadmitted patients, the CC2 and CC3 were 8.2 % (n = 13) and 6.9 % (n = 11). Patients had a median length of hospital stay of 10 days (IQR 8–14 days). A significant difference in length of hospital stay was found between the readmitted and nonreadmitted patients; the median length of stay was 11 days (IQR 9–21 days) for readmitted patients and 10 days (IQR 8–13 days) for nonreadmitted patients (p = 0.02). In-hospital mortality after surgery occurred in 6 patients (3 %). These mortalities were the result of pulmonary embolism (n = 2), multisystem organ failure/cardiopulmonary arrest (n = 1), cardiac dysrhythmia (n = 1), cerebral edema/tonsillar herniation (n = 1), and gastric perforation leading to sepsis (n = 1). Readmission Rates The 30-day readmission rate was 15 % (n = 32) and the 90-day readmission rate was 21 % (n = 46). Reasons for readmission up to 90 days of index hospitalization included infectious complications in 17, including 9 intraabdominal abscesses, 5 wound infections, and 1 case each of urinary tract infection, pyelonephritis, and line infection, as well as abdominal pain (n = 14), malnutrition/failure to thrive (n = 8), bowel obstruction (n = 7), pulmonary embolism/deep-vein thrombosis (n = 3), pleural effusion (n = 2), and hypoglycemia (n = 1). Six patients had more than 1 readmission within 90 days of discharge. The most common causes of readmission in this cohort included abdominal pain (n = 14), dehydration/malnutrition (n = 9), and intra-abdominal abscess (n = 8). A comparison of patient demographics between patients who were readmitted versus those who were not is provided in Tables 1 and 2. There were no significant differences between the 2 groups in terms of gender, race, or age. Early readmissions were those patients readmitted to the hospital within 7 days. Reasons for these early readmissions included abdominal pain (n = 6), intra-abdominal abscess (n = 5), dehydration/malnutrition (n = 5), fever (n = 2), pulmonary embolism/deep-vein thrombosis (n = 2), bowel obstruction (n = 1), and superficial wound infection (n = 1). Patients with ECF and wound infections had a higher risk of readmission. Of the 46 readmitted patients, 6 (13 %) were readmitted on the basis of the presence of an ECF. Three of the 7 patients with ECF were diagnosed at readmission, while the other 4 patients acquired ECF during their index hospitalization. Diagnostic assessment revealed that the ECF originated in the colon in 5 cases, the stomach in 1 case after gastrectomy and gastrojejunostomy, and the

small bowel in 1 case after 3 small bowel resections. There were 6 patients who had more than 1 readmission within 90 days of discharge, and ECF complicated 3 of those 6 readmissions. Surgical site infections were significantly higher among readmitted patients compared to nonreadmitted patients (11.4 % vs. 3.8 % respectively; p = 0.049). Cardiac complications (supraventricular tachycardia, atrial fibrillation, cardiopulmonary arrest), pulmonary complications (pleural effusions, respiratory failure, pneumothorax, pneumonia), intra-abdominal abscess, line infection, and thrombosis were all found not to vary significantly between readmitted and nonreadmitted patients. Survival There was no significant difference in overall survival between the readmitted and nonreadmitted patients. The readmitted and nonreadmitted curves both demonstrated overall survival rates of approximately 60 % at 5 years (Fig. 1). TABLE 1 Patient demographics Characteristic Total Not readmitted (n = 159), n (%)

Readmitted (n = 44), n (%)

Gender Female Male

0.86 113

88 (77.9)

25 (22.1)

90

71 (78.9)

19 (21.1)

188

147 (78.2)

41 (21.8)

Race White Black Other Age at diagnosis 18–40 years

0.58 6

4 (66.7)

2 (33.3)

9

8 (88.9)

1 (11.1) 0.79

40

30 (75.0)

10 (25.0)

40–60 years 109

87 (79.8)

22 (20.2)

60–80 years

38 (76.0)

12 (24.0)

50

[80 years

1

1 (100.0)

0 (0.0)

Unknown

3

3 (100.0)

0 (0.0)

TABLE 2 Site of primary tumor Site*

Total Not readmitted (n = 159), n (%)

Readmitted (n = 44), n (%)

Appendix

163

126 (77.3)

37 (22.7)

15

11 (73.3)

4 (26.7)

Ovarian

7

5 (71.4)

2 (28.6)

Gastric

5

5 (100.0)

0 (0.0)

Mesothelioma

4

3 (75.0)

1 (25.0)

Other

9

9 (100.0)

0 (0.0)

Colon

* p = 0.49

p

A. S. Martin et al.

readmission at 90 days. A multiple logistic regression model was then used to further validate these factors. On multivariate analysis, the only factor that continued to be a significant predictor of readmission was ECF (p = 0.02), with an odds ratio of 14.4 (95 % confidence interval 1.55– 137) (Tables 3, 4).

Overall Survival 1.0 p=0.51

Survival Probability (%)

Readmitted

0.8 Non-readmitted

0.6

0.4

DISCUSSION

0.2

A 2011 report on Medicare patients between 2003 and 2004 by Jencks et al.6 revealed that 19.6 % of 11,855,702 discharged patients were readmitted within 30 days of discharge and that 34 % were rehospitalized after 90 days. Fifty-one percent of patients who were discharged after surgical procedures were readmitted or died within 1 year of discharge, and 70.5 % of all patients readmitted were readmitted for medical conditions. The analysis from this report estimated that the unplanned cost of readmission for both medical and surgical patients to Medicare was $17.4 billion in 2004.6 With these numbers in mind, readmissions for complex surgeries tend to be relatively high, which further drives up the financial burden of rehospitalizations on the healthcare system. More germane to our analyses, complex cancer surgeries including radical cystectomy, pancreaticoduodenectomy, esophagectomy, and lung resection have a high risk of hospital readmission. The 30-day readmission rates for these procedures are reported to range from 20.7 to 25.2,15 to 21, 12.6 to 13.7, and 10 to 12.8 %, respectively.7–14

0.0 0

2

4

6

8

10

12

14

Years

FIG. 1 Overall survival analysis of readmitted patients (n = 159) versus nonreadmitted patients (n = 44) who underwent CRS/HIPEC. Log rank p value for differences in survival between the 2 groups was 0.51. Median survival for nonreadmission was [1917 days. For readmission, median was 2430 days

Factors Associated with Readmission A univariate analysis was then performed to assess for factors predictive readmission after CRS/HIPEC. No associations were found between patient demographics (gender, race, age) and readmission. In addition, tumor characteristics, CC, PCI, and operative factors did not predict need for readmission. Length of hospital stay, presence of ECF, and wound infection predicted TABLE 3 Operative and postoperative characteristics Characteristic

Not readmitted (n = 159)

Readmitted (n = 44)

EBL (ml), median (range)

300 (200–600)

400 (200–600)

p 0.55

Length of stay (day), median (range)

10 (8–13)

11 (9–21)

0.02*

Total PCI, median (range) CC

12 (6–17)

13 (6–18)

0.66 0.058

0

103 (64.8 %)

26 (59.1 %)

1

32 (20.1 %)

12 (27.3 %)

2

13 (8.2 %)

2 (4.6 %)

3

11 (6.9 %)

2 (4.6 %)

0 (0.0 %)

2 (4.6 %)

Unknown Complication Pulmonary

17 (10.8 %)

5 (11.4 %)

0.91

Thrombosis

11 (6.9 %)

3 (6.8 %)

0.98

Wound infection

6 (3.8 %)

5 (11.4 %)

0.049*

IAA

8 (5.0 %)

3 (6.8 %)

0.64

Cardiac

8 (5.0 %)

1 (2.3 %)

0.43

ECF

1 (0.6 %)

6 (13.6 %)

\0.01*

Line infection

2 (1.3 %)

0 (0.0 %)

0.45

EBL estimated blood loss, PCI peritoneal cancer index, CC completeness of cytoreduction, EC enterocutaneous, IAA intra-abdominal abscess * p \ 0.05

Factors Associated with Readmission After Cytoreductive Surgery… TABLE 4 Predictors of readmission Characteristic

Univariate analysis OR (95 % CI)

Female

1.06 (0.54–2.08)

Race

p

OR (95 % CI)

p

1.03 (0.99–1.06)

0.14

1.90 (0.46–7.90)

0.38

14.5 (1.55–137)

0.02*

0.86 0.60

White

1.00

Black

1.79 (0.32–10.1)

Other

0.45 (0.05–3.69)

Age 18–40 years 40–60 years

Multivariate analysis

0.95 1.00 0.75 (0.32–1.75)

60–80 years

0.94 (0.36–2.46)

[80 years

1.05 (0.01–94.0)

Unknown

0.42 (0.01–13.7)

Site of primary lesion

0.79

Appendix

1.00

Colon

1.32 (0.40–4.31)

Ovarian

1.53 (0.30–7.90)

Gastric

0.31 (0.01–7.47)

Mesothelioma

1.45 (0.17–12.7)

Other

0.18 (0.01–3.64)

EBL (ml)

1.00 (1.00–1.00)

0.67

Length of stay (day)

1.04 (1.01–1.07)

\0.01*

Pulmonary Thrombosis

1.06 (0.37–3.06) 0.99 (0.26–3.70)

0.91 0.98

Wound infection

3.27 (0.95–11.3)

0.06

IAA

1.38 (0.35–5.44)

0.64

Cardiac

0.44 (0.05–3.61)

0.44

ECF

24.9 (2.92–213)

\0.01*

Line infection

0.71 (0.02–29.6)

0.86

1.01 (0.96–1.06)

0.72

Complications

Total PCI CC

0.43

0

1.00

1

1.50 (0.68–3.30)

2

0.72 (0.17–3.11)

3

0.85 (0.19–3.74)

Unknown

0.96 (0.46–827)

EBL estimated blood loss, IAA intra-abdominal abscess, ECF enterocutaneous fistula, PCI peritoneal cancer index, CC completeness of cytoreduction * p \ 0.05

CRS/HIPEC represents one of the most complex operations performed by surgeons. Recent literature has shown improved outcomes in patients who underwent a complete resection via CRS/HIPEC for pseudomyxoma peritonei and other histology associated with peritonealbased metastases. Survival improved from 12.6 to 22.3 months with CRS/HIPEC compared to systemic

chemotherapy alone in a landmark article by Verwaal et al.15 Despite the improvement in survival, significant morbidity can occur after this operation. In a separate study, Verwaal et al. specifically evaluated the morbidity after CRS/HIPEC, reporting morbidity rates of 35 %, with the majority experiencing grade 3–5 toxicity.

A. S. Martin et al.

It is surprising, therefore, that there have been only a few reports on readmission rates after CRS/HIPEC. An analysis of the American College of Surgeons National Surgical Quality Improvement Program reported 30-day readmission rates after CRS/HIPEC at 11 %, with an average length of stay of 13 days for the 694 patients sampled, 97 of which were appendiceal in pathology.16 Several other studies demonstrate 30-day readmission rates after CRS/HIPEC to be approximately 11–37 %; none of these studies specifically addressed the reasons for their 30day readmissions (Table 5; Supplemental Material). Our study is the first of its kind to delineate factors associated with readmission after CRS/HIPEC. In the current study, we assessed over 200 CRS/HIPEC cases with peritoneal malignancy in an attempt to analyze 30- and 90-day readmission rates as well as the factors contributing to readmission. Limiting and assessing readmissions within 90 days helped ensure that readmissions were related to surgery alone and were not the result of disease progression or concomitant morbidities. Our 30and 90-day readmission rates of 15 and 21 % fall within previously reported readmission rates (Table 5). The most common reasons for readmission included abdominal pain (n = 14), intra-abdominal abscess (n = 9), malnutrition/failure to thrive (n = 8), and bowel obstruction (n = 7). Another important finding in our study was that 3 % of patients required multiple readmissions. Univariate analysis at 90 days showed length of hospital stay, ECF complication, and wound infection as being predictive of readmission. However, on subsequent multivariate analysis, only ECF remained a predictor of readmission. ECF was diagnosed during the index hospitalization in 3 of 7 patients who were readmitted, and the rest were diagnosed in the remaining patients during their readmission. Our findings are consistent with other previous reports in which enteric fistulas have been shown to be one of the most common complications after CRS and HIPEC.17 A smaller

10-year study (2005–2014) of 184 patients with peritoneal carcinomatosis undergoing CRS/HIPEC showed a similar rate of 8.6 % (n = 16) for ECF.18,19 Just as important, factors not associated with higher readmission rates included extent of disease, extent of cytoreduction, blood loss, and infectious complications including pulmonary complications, line infections, wound infections, and intraabdominal abscesses. PCI and CC were not associated with higher readmission rates in this study despite being previously reported as indicators of increased morbidity and mortality.20 This also suggests that patients undergoing aggressive surgery are not more likely to be readmitted. This is further supported by the fact that patients undergoing pancreatectomy and liver resections were not more likely to be readmitted. This may indicate that complications that occur as a result of more aggressive surgery such as wound infections, pneumonia, and intra-abdominal abscesses are being recognized and treated during the index stay. Similar to other published CRS/HIPEC series of this nature, the main limitation of our study is its retrospective nature. As with other studies evaluating readmissions, we cannot be certain that all hospital readmissions were captured in our study. Thus, it is possible that our findings underrepresent the true readmission rate. Of particular interest to us were patients being readmitted within 1 week of discharge. Presumably these patients represent a cohort of patients whose readmission may have been avoidable. In our experience, 21 patients (46 % of all readmitted) were readmitted within 1 week of discharge. The reasons for readmission within 1 week of discharge included poorly managed pain, poor nutrition, and infectious complications. One could argue that many of these reasons could have been addressed before discharge and that perhaps these patients were discharged too early. If readmissions can be reduced, this subset of patients may represent a group where monitoring early signs of infection, addressing and

TABLE 5 Summary of recently published data regarding readmission rates after CRS/HIPEC Characteristic

Votanopoulos et al.21 Jafari et al.16

Austin et al.22

Randle et al.23

Year

2013

2014

2013

2015

Cohort

246 patients, 272 cases

694 patients

282 patients

974 patients, 1067 cases

Readmission rates (30 days)

19–25 %

11 %

13.8 %, 39 readmitted

27.8–37 %

Appendiceal primary tumor

49.6 % (135 cases)

14 %

100 %; DPAM 25.2 %; PMCA 74.5 %

46 %

ECF

4 (1.5 %)

65 (9.4 %) both anastomotic leaks and ECF

6 (2.1 %)

–

CRS cytoreductive surgery, HIPEC hyperthermic intraperitoneal chemotherapy, SSI surgical site infection, DPAM disseminated peritoneal adenomucinosis, PMCA peritoneal mucinous carcinomatosis, ECF enterocutaneous fistula

Factors Associated with Readmission After Cytoreductive Surgery…

properly treating nutritional requirements, and providing pain control may avert the need for subsequent early readmissions. In summary, our data set represents an analysis of factors attributed to readmission after CRS/HIPEC in patients with pseudomyxoma peritonei. Although we found that a significant risk for readmission remains after CRS/HIPEC, with 30- and 90-day readmission rates of 15 and 24 %, our readmission rates are in line with previous HIPEC reports as well as other complex cancer surgeries. Many readmissions after complex surgery may be unavoidable, and its use as a quality metric continues to be debated. It is crucial, therefore, to establish baseline rates of readmission at high-volume HIPEC centers and to determine what represent acceptable readmission rates. Here we have established a benchmark number for acceptable readmission rates after CRS/HIPEC, thus providing a baseline from which other centers can evaluate their own readmission rates and surgical complications. DISCLOSURE

The authors declare no conflict of interest.

REFERENCES 1. Stitzenberg KB, Chang Y, Smith AB, Nielsen ME. Exploring the burden of inpatient admissions after major cancer surgery. J Clin Oncol. 2015;33:455–64. 2. Ahmad SA, Kim J, Sussman JJ, et al. Reduced morbidity following cytoreductive surgery and intraperitoneal hyperthermic chemoperfusion. Ann Surg Oncol. 2004;11:387–92. 3. 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. 1999;6:790–6. 4. Sugarbaker PH. Peritonectomy procedures. Ann Surg. 1995;221:29–41. 5. Verwaal VJ, van Tinteren H, Ruth SV, et al. Toxicity of cytoreductive surgery and hyperthermic intra-peritoneal chemotherapy. J Surg Oncol. 2004; 85:61–7. 6. Jencks SF, Williams MV, Coleman EA, et al. Rehospitalizations among patients in the medicare fee-for-service program. N Engl J Med. 2009;360:1418–28. 7. Jacobs BL, Zhang Y, Tan HJ, Ye Z, Skolarus TA, Hollenbeck BK. Hospitalization trends after prostate and bladder surgery: implications of potential payment reforms. J Urol. 2013;189:59– 65. 8. Leow JJ, Gandaglia G, Sood A, et al. Readmissions after major urologic cancer surgery. Can J Urol. 2014;21:7537–46. 9. Ahmad SA, Edwards MJ, Sutton JM, et al. Factors influencing readmission after pancreaticoduodenectomy: a multi-institutional study of 1302 patients. Ann Surg. 2012;256:529–37.

10. Hyder O, Dodson RM, Nathan H, et al. Influence of patient, physician, and hospital factors on 30-day readmission following pancreatoduodenectomy in the United States. JAMA Surg. 2013;148:1095–102. 11. Shah SP, Xu T, Hooker CM, et al. Why are patients being readmitted after surgery for esophageal cancer? J Thorac Cardiovasc Surg. 2015;149:1384–9. 12. Sundaram A, Srinivasan A, Baker S, Mittal SK. Readmission and risk factors for readmission following esophagectomy for esophageal cancer. J Gastrointest Surg. 2015;19:581–5. 13. Hu Y, McMurry TL, Isbell JM, Stukenborg GJ, Kozower BD. Readmission after lung cancer resection is associated with a 6fold increase in 90-day postoperative mortality. J Thorac Cardiovasc Surg. 2014;148:2261–7. 14. McDevitt J, Kelly M, Comber H, et al. A population-based study of hospital length of stay and emergency readmission following surgery for non-small-cell lung cancer. Eur J Cardiothorac Surg. 2013;44:e253–9. 15. Verwaal VJ, van Ruth S, de Bree E, et al. Randomized trial of cytoreduction and hyperthemic intraperitoneal versus systemic chemotherapy and palliative surgery in patients with peritoneal carcinomatosis of colorectal cancer. J Clin Oncol. 2003;21:3737– 43. 16. Jafari MD, Halabi WJ, Stamos MJ, et al. Surgical outcomes of hyperthermic intraperitoneal chemotherapy: analysis of the American College of Surgeons National Surgical Quality Improvement Program. Pacific Coast Surgical Association. JAMA. 2014;149:170–5. 17. Glehen O, Gilly FN, Boutitie F, et al. French Surgical Association. Toward curative treatment of peritoneal carcinomatosis from nonovarian origin by cytoreductive surgery combined with perioperative intraperitoneal chemotherapy: a multi-institutional study of 1290 patients. Cancer. 2010;116:5608–18. 18. Halkia E, Efstathiou E, Rogdakis A, et al. Digestive fistulas after cytoreductive surgery and HIPEC in peritoneal carcinomatosis. J BUON. 2015;20(1 suppl):S60–3. 19. Haslinger M, Francesecutti V, Attwood K, et al. A contemporary analysis of morbidity and outcomes in cytoreduction/hyperthermic intraperitoneal chemoperfusion. Cancer Med. 2013;2:334– 42. 20. da Silva RG, Sugarbaker PH. Analysis of prognostic factors in seventy patients having a complete cytoreduction plus perioperative intraperitoneal chemotherapy for carcinomatosis from colorectal cancer. J Am Coll Surg. 2006;203:878–86. 21. Votanopoulos KI, Swords DS, Swett KR, et al. Obesity and peritoneal surface disease: outcomes after cytoreductive surgery with hyperthermic intraperitoneal chemotherapy for appendiceal and colon primary tumors. Ann Surg Oncol. 2013;20:3899–904. 22. Austin F, Mavanur A, Sathaiah M, et al. Aggressive management of peritoneal carcinomatosis from mucinous appendiceal neoplasms. Ann Surg Oncol. 2013;19:1386–93. 23. Randle RW, Doud AN, Levine EA, et al. Peritoneal surface disease with synchronous hepatic involvement treated with cytoreductive surgery (CRS) and hyperthermic intraperitoneal chemotherapy (HIPEC). Ann Surg Oncol. 2015;22:1634–8.