J Gastrointest Surg DOI 10.1007/s11605-014-2559-4
ORIGINAL ARTICLE
Prognostic Factors for Postoperative Morbidity and Tumour Response After Neoadjuvant Chemoradiation Followed by Resection for Rectal Cancer Annefleur E. M. Berkel & Dankert P. Woutersen & Job van der Palen & Joost M. Klaase
Received: 12 April 2014 / Accepted: 27 May 2014 # 2014 The Society for Surgery of the Alimentary Tract
Abstract Background and Purpose In patients with locally advanced rectal cancer treated with neoadjuvant chemoradiation followed by rectal resection, postoperative morbidity is a significant clinical problem. Pathologic complete tumour response seems to give the best prognosis in the long term. Little is known about the factors that are associated with postoperative complications and pathologic complete response. The aim of this retrospective study was to identify and describe these factors. Methods Ninety-nine consecutive patients with locally advanced rectal cancer who underwent neoadjuvant chemoradiation (50 Gy and capecitabine) followed by surgery at our institute between January 2007 and May 2012 were identified. Postoperative complications were graded according to the Clavien-Dindo classification. Pathologic tumour response was categorized as complete response or no/partial response. Results Postoperative complications occurred in 68 patients (69 %) and grade 3–5 complications in 25 patients (25 %). The 30day and 90-day mortality were 1 % (n=1) and 2 % (n=2), respectively. A young age (p=0.021) and a preoperative or postoperative blood transfusion (p=0.015) independently predicted complications. Intraoperative or postoperative blood transfusion (p=0.007) and ypT0-1 stage (p=0.037) were independent predictors for grade 3–5 complications. Complete response rate was 22 % (n=22); 4 % (n=4) of patients showed no response. No independent factors predicting complete response were found. Conclusions Neoadjuvant chemoradiation followed by rectal resection is associated with significant postoperative morbidity but minimal postoperative mortality. A complete response rate of 22 % was achieved.
The paper was presented at the ‘Chirurgendagen’ 30 mei 2013, Veldhoven, The Netherlands. A. E. M. Berkel : J. M. Klaase (*) Department of Surgery, Medisch Spectrum Twente, Haaksbergerstraat 55, 7500, KA Enschede, The Netherlands e-mail: [email protected] A. E. M. Berkel e-mail: [email protected] D. P. Woutersen Department of Radiotherapy, Medisch Spectrum Twente, Enschede, The Netherlands J. van der Palen Department of Epidemiology, Medisch Spectrum Twente, Enschede, The Netherlands J. van der Palen Department of Research Methodology, Measurement and Data Analysis, University of Twente, Enschede, The Netherlands
Keywords Rectal cancer . Neoadjuvant . Chemoradiation . Postoperative complications . Tumour response
Introduction Postoperative Complications In our hospital, 43 % of patients with colorectal cancer have a locally advanced carcinoma.1 According to the Dutch guidelines, treatment of locally advanced rectal cancer consists of neoadjuvant chemoradiation followed by rectal surgery.2 This extensive treatment regimen often results in postoperative complications. The postoperative complication rate is about 40 %.3, 4 Especially, anastomotic leakage following low anterior resection (LAR) and perineal wound complications following abdominoperineal resection (APR) are important and frequent complications, with an average of 10 %5 and 35 % respectively.6, 7 Kerr et al (2008)6 showed that a shorter
J Gastrointest Surg
interval between chemoradiation and rectal surgery significantly predicted more anastomotic leakages and perineal wound complications. Stelzmueller et al (2009)4 showed that women, low pretherapeutic haemoglobin level and higher ASA score were associated with early postoperative complications. Also, elderly patients are prone to complications, especially general complications, because of their comorbidities and the physiologic factors that play a role in aging.5 These complications often lead to a prolonged hospitalization and also to a higher mortality rate.5 Identifying factors associated with postoperative complications seem important, because postoperative complications are associated with worse long-term survival rates.4 Tumour Response After Chemoradiation Neoadjuvant chemoradiation for locally advanced rectal cancer has been shown to achieve pathologic tumour downstaging, with improvement of resectability and sphincter saving surgery8 and has been shown to reduce the risk of developing local recurrences or distance recurrences.9 Pathologic complete response (pCR) seems to translate into improved clinical results. There is a large variability in response to neoadjuvant chemoradiation between patients. Some patients do not respond at all and may even have progression of their disease, while others have a pCR. Incidence of pCR ranges from 10 to 24 %.3, 4, 6, 10–13 Several studies showed that a longer interval between chemoradiation and surgery significantly improves the rate of pCR.3, 10, 11, 14–16 It remains unknown what the optimal time interval between chemoradiation and surgery is. Increased intervals could potentially increase the tumour downstaging effect because radiation-induced necrosis appears to be a timedependent phenomenon, but a longer interval can also allow the tumour to progress and decrease survival.17 Das et al (2008)12 showed that a circumferential extent of the tumour >60 %, carcinoembryonic antigen (CEA) level >2.5 ng/ml and tumour distance from the anal verge >5 cm were associated significantly with a lower pCR rate. Identifying factors that are associated with pCR could help clinicians with treatment decisions and determining prognosis. Few studies investigated these factors, so it is unknown which patients with locally advanced rectal cancer benefit the most from treatment with neoadjuvant chemoradiation. The aim of the present study was to identify prognostic factors for postoperative complications and pCR in patients with locally advanced rectal cancer treated with neoadjuvant chemoradiation followed by surgery.
Materials and Methods All consecutive patients between January 2007 and May 2012 with locally advanced rectal cancer who had neoadjuvant
chemoradiation and surgery in Medical Spectrum Twente, a large non-academic teaching hospital in the Netherlands, were retrospectively identified. Locally advanced rectal cancer was defined as a clinical T4 tumour or T3 tumour with a threatened circumferential resection margin (CRM6 weeks persisting skin defect) following APR Any fluid collections around the rectal stump/presacral after a Hartmann procedure Wound dehiscence, wound infections, wound necrosis, abscess or delayed wound healing (>6 weeks persisting skin defect) Any intra-abdominal fluid collections unrelated to the anastomosis or perineal wound Enterocutaneous, enterovaginal, vesicovaginal or enterovesical connections Systemic inflammatory response syndrome caused by an infection Absence of bowel sounds or defecation more than 5 days following surgery; need for placement or delayed removal of nasogastric tube Gastrointestinal haemorrhage, decrease in haemoglobin level directly after surgery treated conservatively with or without blood transfusions or by reintervention Any bowel ischaemia Problems with stoma, such as parastomal hernia or stenosis Ureter leakage, urinary incontinence, ureter stenosis, suprapubic or transurethrale catheter-related problems, urinary tract infections, urinary retention, urosepsis Sexual problem such as retrograde ejaculation, secondary dyspareunia
Perineal wound complication Rectal stump abscess Abdominal wound complication Intra-abdominal abscess Fistula Sepsis Ileus Bleeding
Intestinal necrosis Stoma complication Urological LAR low anterior resection, CT computed tomography, Hb haemoglobin, APR abdominoperineal resection
Sexual dysfunction
The severity was scored using the Clavien-Dindo classification of surgical complications (grades 1–5).18, 19 Grade 1 complications are complications without the need for treatment (except medication as antiemetics, and wound infections opened at bedside). Grade 2 complications require pharmacological treatment. Grade 3 complications require surgical, endoscopic or radiological intervention. Grade 4 complications are life-threatening complications requiring IC/ICU management. Grade 5 is death of a patient. General complications were scored as cardiovascular, pulmonary, renal or neurological. Pathologic tumour response was divided into complete response and no/partial response. Pathologic complete response was defined as absence of viable adenocarcinoma cells in the surgical specimen, including primary tumour
and lymph nodes. Patients were dichotomized into two groups according to the interval between neoadjuvant chemoradiation and surgery, following a proposal published in the literature: ≤7 weeks (short-interval group) and >7 weeks (long-interval group).3, 11 Statistical Analysis Data was analysed with SPSS 20.0. Data are presented with median and range or mean and standard deviation, as appropriate. Categorical data are summarized by frequency and percentage within each cohort. The univariate associations between variables and postoperative (grades 3–5) complications and pathologic complete response were evaluated for
Fig. 1 Flow of treatment with number of patients (n) and reasons of exclusion. TME total mesorectal excision
Neoadjuvant chemoradiation n = 105
Pre-operative exclusion n= 2 New metastases n= 1 Comorbidity n = 1 Laparotomy n = 103 Perioperative exclusion n= 4 Unresectable n= 4
Rectal resection (TME) n = 99
Low anterior resection n= 18
Hartmann procedure n= 22 Total exenteration n= 1
Abdominoperineal resection n= 59 Posterior exenteration n= 1 Total exenteration n= 1
J Gastrointest Surg
significance, using the Fisher’s exact test or Chi-square test for categorical variables and the Mann–Whitney U test or independent t test (if normally distributed) for continuous variables. If the number of persons per category was small, categories were merged. Variables with a p
ORIGINAL ARTICLE
Prognostic Factors for Postoperative Morbidity and Tumour Response After Neoadjuvant Chemoradiation Followed by Resection for Rectal Cancer Annefleur E. M. Berkel & Dankert P. Woutersen & Job van der Palen & Joost M. Klaase
Received: 12 April 2014 / Accepted: 27 May 2014 # 2014 The Society for Surgery of the Alimentary Tract
Abstract Background and Purpose In patients with locally advanced rectal cancer treated with neoadjuvant chemoradiation followed by rectal resection, postoperative morbidity is a significant clinical problem. Pathologic complete tumour response seems to give the best prognosis in the long term. Little is known about the factors that are associated with postoperative complications and pathologic complete response. The aim of this retrospective study was to identify and describe these factors. Methods Ninety-nine consecutive patients with locally advanced rectal cancer who underwent neoadjuvant chemoradiation (50 Gy and capecitabine) followed by surgery at our institute between January 2007 and May 2012 were identified. Postoperative complications were graded according to the Clavien-Dindo classification. Pathologic tumour response was categorized as complete response or no/partial response. Results Postoperative complications occurred in 68 patients (69 %) and grade 3–5 complications in 25 patients (25 %). The 30day and 90-day mortality were 1 % (n=1) and 2 % (n=2), respectively. A young age (p=0.021) and a preoperative or postoperative blood transfusion (p=0.015) independently predicted complications. Intraoperative or postoperative blood transfusion (p=0.007) and ypT0-1 stage (p=0.037) were independent predictors for grade 3–5 complications. Complete response rate was 22 % (n=22); 4 % (n=4) of patients showed no response. No independent factors predicting complete response were found. Conclusions Neoadjuvant chemoradiation followed by rectal resection is associated with significant postoperative morbidity but minimal postoperative mortality. A complete response rate of 22 % was achieved.
The paper was presented at the ‘Chirurgendagen’ 30 mei 2013, Veldhoven, The Netherlands. A. E. M. Berkel : J. M. Klaase (*) Department of Surgery, Medisch Spectrum Twente, Haaksbergerstraat 55, 7500, KA Enschede, The Netherlands e-mail: [email protected] A. E. M. Berkel e-mail: [email protected] D. P. Woutersen Department of Radiotherapy, Medisch Spectrum Twente, Enschede, The Netherlands J. van der Palen Department of Epidemiology, Medisch Spectrum Twente, Enschede, The Netherlands J. van der Palen Department of Research Methodology, Measurement and Data Analysis, University of Twente, Enschede, The Netherlands
Keywords Rectal cancer . Neoadjuvant . Chemoradiation . Postoperative complications . Tumour response
Introduction Postoperative Complications In our hospital, 43 % of patients with colorectal cancer have a locally advanced carcinoma.1 According to the Dutch guidelines, treatment of locally advanced rectal cancer consists of neoadjuvant chemoradiation followed by rectal surgery.2 This extensive treatment regimen often results in postoperative complications. The postoperative complication rate is about 40 %.3, 4 Especially, anastomotic leakage following low anterior resection (LAR) and perineal wound complications following abdominoperineal resection (APR) are important and frequent complications, with an average of 10 %5 and 35 % respectively.6, 7 Kerr et al (2008)6 showed that a shorter
J Gastrointest Surg
interval between chemoradiation and rectal surgery significantly predicted more anastomotic leakages and perineal wound complications. Stelzmueller et al (2009)4 showed that women, low pretherapeutic haemoglobin level and higher ASA score were associated with early postoperative complications. Also, elderly patients are prone to complications, especially general complications, because of their comorbidities and the physiologic factors that play a role in aging.5 These complications often lead to a prolonged hospitalization and also to a higher mortality rate.5 Identifying factors associated with postoperative complications seem important, because postoperative complications are associated with worse long-term survival rates.4 Tumour Response After Chemoradiation Neoadjuvant chemoradiation for locally advanced rectal cancer has been shown to achieve pathologic tumour downstaging, with improvement of resectability and sphincter saving surgery8 and has been shown to reduce the risk of developing local recurrences or distance recurrences.9 Pathologic complete response (pCR) seems to translate into improved clinical results. There is a large variability in response to neoadjuvant chemoradiation between patients. Some patients do not respond at all and may even have progression of their disease, while others have a pCR. Incidence of pCR ranges from 10 to 24 %.3, 4, 6, 10–13 Several studies showed that a longer interval between chemoradiation and surgery significantly improves the rate of pCR.3, 10, 11, 14–16 It remains unknown what the optimal time interval between chemoradiation and surgery is. Increased intervals could potentially increase the tumour downstaging effect because radiation-induced necrosis appears to be a timedependent phenomenon, but a longer interval can also allow the tumour to progress and decrease survival.17 Das et al (2008)12 showed that a circumferential extent of the tumour >60 %, carcinoembryonic antigen (CEA) level >2.5 ng/ml and tumour distance from the anal verge >5 cm were associated significantly with a lower pCR rate. Identifying factors that are associated with pCR could help clinicians with treatment decisions and determining prognosis. Few studies investigated these factors, so it is unknown which patients with locally advanced rectal cancer benefit the most from treatment with neoadjuvant chemoradiation. The aim of the present study was to identify prognostic factors for postoperative complications and pCR in patients with locally advanced rectal cancer treated with neoadjuvant chemoradiation followed by surgery.
Materials and Methods All consecutive patients between January 2007 and May 2012 with locally advanced rectal cancer who had neoadjuvant
chemoradiation and surgery in Medical Spectrum Twente, a large non-academic teaching hospital in the Netherlands, were retrospectively identified. Locally advanced rectal cancer was defined as a clinical T4 tumour or T3 tumour with a threatened circumferential resection margin (CRM6 weeks persisting skin defect) following APR Any fluid collections around the rectal stump/presacral after a Hartmann procedure Wound dehiscence, wound infections, wound necrosis, abscess or delayed wound healing (>6 weeks persisting skin defect) Any intra-abdominal fluid collections unrelated to the anastomosis or perineal wound Enterocutaneous, enterovaginal, vesicovaginal or enterovesical connections Systemic inflammatory response syndrome caused by an infection Absence of bowel sounds or defecation more than 5 days following surgery; need for placement or delayed removal of nasogastric tube Gastrointestinal haemorrhage, decrease in haemoglobin level directly after surgery treated conservatively with or without blood transfusions or by reintervention Any bowel ischaemia Problems with stoma, such as parastomal hernia or stenosis Ureter leakage, urinary incontinence, ureter stenosis, suprapubic or transurethrale catheter-related problems, urinary tract infections, urinary retention, urosepsis Sexual problem such as retrograde ejaculation, secondary dyspareunia
Perineal wound complication Rectal stump abscess Abdominal wound complication Intra-abdominal abscess Fistula Sepsis Ileus Bleeding
Intestinal necrosis Stoma complication Urological LAR low anterior resection, CT computed tomography, Hb haemoglobin, APR abdominoperineal resection
Sexual dysfunction
The severity was scored using the Clavien-Dindo classification of surgical complications (grades 1–5).18, 19 Grade 1 complications are complications without the need for treatment (except medication as antiemetics, and wound infections opened at bedside). Grade 2 complications require pharmacological treatment. Grade 3 complications require surgical, endoscopic or radiological intervention. Grade 4 complications are life-threatening complications requiring IC/ICU management. Grade 5 is death of a patient. General complications were scored as cardiovascular, pulmonary, renal or neurological. Pathologic tumour response was divided into complete response and no/partial response. Pathologic complete response was defined as absence of viable adenocarcinoma cells in the surgical specimen, including primary tumour
and lymph nodes. Patients were dichotomized into two groups according to the interval between neoadjuvant chemoradiation and surgery, following a proposal published in the literature: ≤7 weeks (short-interval group) and >7 weeks (long-interval group).3, 11 Statistical Analysis Data was analysed with SPSS 20.0. Data are presented with median and range or mean and standard deviation, as appropriate. Categorical data are summarized by frequency and percentage within each cohort. The univariate associations between variables and postoperative (grades 3–5) complications and pathologic complete response were evaluated for
Fig. 1 Flow of treatment with number of patients (n) and reasons of exclusion. TME total mesorectal excision
Neoadjuvant chemoradiation n = 105
Pre-operative exclusion n= 2 New metastases n= 1 Comorbidity n = 1 Laparotomy n = 103 Perioperative exclusion n= 4 Unresectable n= 4
Rectal resection (TME) n = 99
Low anterior resection n= 18
Hartmann procedure n= 22 Total exenteration n= 1
Abdominoperineal resection n= 59 Posterior exenteration n= 1 Total exenteration n= 1
J Gastrointest Surg
significance, using the Fisher’s exact test or Chi-square test for categorical variables and the Mann–Whitney U test or independent t test (if normally distributed) for continuous variables. If the number of persons per category was small, categories were merged. Variables with a p
