RESEARCH PROJECT OBJECTIVES The aim of this planned study is to evaluate factors, with particular reference to surgical technique of abdominal closure, accumulation of advanced glycation end products, and collagen content in the transversalis fascia, that are associated with the development of incisional hernias after transverse epigastric incisions in patients operated due to malignant tumors.

RESEARCH PROJECT METHODOLOGY:

The study is designed as prospective and is planned to include 392 patients undergoing abdominal surgery due to malignant tumors of the alimentary system performed through transverse incisions in the epigastrium. Primary end-point of the study is defined as the occurrence of burst abdomen during immediate postoperative period or incisional hernia over 2 year postoperative follow-up. Assessment of the association between the type of surgical technique and development of hernia will be based on comparison of mass (1 layer) continuous suture and layered (2 layers) continuous suture using slowly absorbable material and applying suture to wound length ratio of more than 4. Type of the utilized surgical technique will be based on randomization. The method of abdominal closure will remain unknown for patients and investigators assessing the presence of incisional hernias. Assessment of overall collagen content and type I to type III ratio will be performed by obtaining a fragment of tissue during operation, preparation of formalin-fixed and paraffin embedded blocks, cutting of 4-micrometer thick sections, staining with picrosirius red and immunohistochemical procedures. Images will be analyzed with dedicated computer software. Accumulation of advanced glycation end products will be evaluated indirectly by measuring skin autofluorescence utilizing a method based on the use of photodiodes. Postoperative follow-up will include the period of postoperative hospitalization and two additional control visits at 1 and 2 years after the operation. Assessment of the presence of incisional hernia will comprise clinical examination, ultrasonographical study, and analysis of images from other available radiological studies. Irrespective of the primary end-point, additional analyses will be performed concerning associations between evaluated factors and occurrence of burst abdomen and the impact of incisional hernia on patients quality of life using the EORTC QLQ C-30 questionnaires. Statistical analyses will, among other, include Kaplan-Meier method, log-rank test, Cox proportional hazards regression and logistic regression

RESEARCH PROJECT OBJECTIVES The aim of the presented scientific project is to establish the factors, with special reference to the type of surgical technique, tissue accumulation of advanced glycation end products, and collagen content in transversalis fascia, associated with development of incisional hernia after transverse incision in the epigastric region in patients operated for malignant tumors WORK PLAN This study is planned as a prospective evaluation of factors associated with the development of incisional hernias after transverse incisions in the epigastric region in patients undergoing surgical treatment of malignant tumors of the alimentary system. The main hypothesis of this study, which forms the basis for sample size calculations, is the presence of an association between application of mass continuous abdominal closure and reduction of the risk of incisional hernia from 35% down to 20% as compared to layered continuous suture. Using a threshold for the level of type I error of 0.05 and a threshold for the probability of type II error on the level of 20%, the total number of patients who should be included in the study is 284 (142 patients in either of the 2 arms). However, considering the approximate value of 2-year patient survival of 20% and a further 10% rate of losses to follow-up, the study is planned to include a cumulative number of 392 patients (196 in either of the two arms).

The primary end-point of this study is the development of burst abdomen in the immediate postoperative period (until discharge from hospital) or incisional hernia over a 2-year follow-up period (combined end-point). The follow-up protocol used for the assessment of incisional hernias at discharge from hospital and two control visits at 1 and 2 years postoperatively. Evaluation of the presence of incisional hernia will, in each case, be based upon both clinical and ultrasonographical examination.

The primary factor that will be assessed for potential association with the development of incisional hernias is the surgical technique applied for abdominal closure. The technique comprising continuous suture and mass closure (one layer of sutures involving peritoneum, transversalis fascia, posterior and anterior layer of rectus abdominis muscle fascia and potentially, fascia of the oblique abdominal muscles) will be compared to the technique 2 separate layers of continuous sutures involving peritoneum, transversalis fascia, and posterior layer of rectus abdominis muscle fascia (first layer), and anterior layer of rectus abdominis muscle fascia and potentially, with oblique abdominal muscles fascia (second layer), respectively. Patients will be randomly assigned to one of the two arms in a 1:1 ratio basing on drawing a sealed envelope with code specific for the type of intervention. Neither the patients included in the study nor the investigators assessing the presence of incisional hernia during follow-up visits will be aware of the type of surgical technique used for abdominal closure (double-blinded study).

Only the professionals present in the operating room during surgical procedure, including operator, assistants, anesthesiologist, and nurses, will be aware of the type of assigned intervention. Besides the type of surgical technique and independent of random assignment to one of the two arms of the study, all included patients will be subject to the assessment of accumulation of advanced glycation end products within the skin and collagen in general, type I collagen and type III collagen within the transversalis fascia in order to evaluate potential associations between these factors and occurrence of the primary end-point. Analyses on the influence of the type of surgical technique used for abdominal closure, accumulation of advanced glycation end products, and collagen content within the transversalis fascia on the risk of incisional hernia formation will be adjusted for the effects of remaining known risk factors, including: age, sex, weight, height, primary disease being an indication for the procedure, history of previous hernia, history of previous abdominal operations, preoperative chemotherapy, cigarette smoking, excessive alcohol intake, presence of liver disease, obesity, arterial hypertension, chronic obstructive pulmonary disease, heart failure, anemia, race, width of subcutaneous adipose tissue, particular type and length of the incision, intraoperative transfusions of packed red blood cells, and occurrence of surgical site infection in the postoperative period.

Analyses concerning the primary end-point of the study will be performed 2 years after the date of surgical procedure in the last patient included in the study cohort. Additionally, an early analysis only with respect to the occurrence of burst abdomen is planned to be done 30 days after the date of surgical procedure in the last patient included in the study cohort. Independently of the primary end-point, separate analyses will be performed to evaluate the impact of the analyzed factors and the occurrence of incisional hernias on the quality of patients' life.

RESEARCH METHODOLOGY Recruitment of patients into the planned research project will be performed on the basis of strict inclusion and exclusion criteria. Inclusion criteria will comprise: age over 18 years, scheduled operative treatment due to malignant tumor of the alimentary system (including liver, bile ducts, and pancreas) through transverse incision in the epigastric region, and provision of informed consent to participate in the study. Exclusion criteria will comprise: necessity to perform an urgent operation, a history of previous surgery performed with transverse incision in the epigastric region, and a body mass index >35 kg/m2.

Following inclusion of patient in the study, baseline anthropometric measurements will be performed (height, weight, and waist circumference), along with physical examination and collection of patient's medical history. All medical documents and additional laboratory, radiological and any other type of findings will be analyzed and relevant data collected in order to provide information on the previously established risk factors for the development of incisional hernia. All the data will be gathered both as paper documentation and as records in a dedicated electronic database. In the next step, accumulation of advanced glycation end products in the skin of the forearm will be evaluated using an indirect measurement method based on the assessment of skin autofluorescence. This phenomenon is a characteristic finding caused by the presence of several advanced glycation end products and its quantitative assessment within the skin was found to precisely reflect the magnitude of accumulation of advanced glycation and products in general in other tissues, which was previously confirmed by comparing skin autofluorescence findings with the results of tissue biopsies [29]. Due to its non-invasive character and reliable results, evaluation of advanced glycation end products accumulation using the method of quantitative assessment of skin autofluorescence was utilized in numerous clinical studies worldwide. In the planned research project, quantitative measurement of skin autofluorescence will be performed with a non-invasive device, which use was previously validated in a clinical setting (AGE-reader connect, www.diagnoptics.com, DiagnOptics BV, Groningen, The Netherlands). Accordingly, skin autofluorescence will be excited by emission of ultraviolet (UV-A) light with a wavelength of 375 nm (E=7.81E-01 Wm2 @ 0.2m). Light emitted in response by the skin will be measured using a system based on photodiodes. Calculation of arbitrary units is performed by a dedicated software of the device. Three separate 12-second measurements will be performed in each patient on the anterior side of the forearm approximately 10 cm below the elbow fold to decrease the potential risk of measurement error. The final skin autofluorescence value will be the mean score of these 3 separate measurements. Importantly, this parameter will only be assessed in patients with skin phototypes I, II, and III according to the Fitzpatrick classification. Skin autofluorescence will also be assessed during the two postoperative follow-up visits at 1 and 2 years.

In the preoperative period, all the patients will be subject to a baseline quality of life assessment. Therefore, all patients will be asked to fill the polish version of the quality of life questionnaire core 30 (QLQ-C30) form prepared by the European Organization for Research and Treatment of Cancer (EORTC). The EORTC QLQ-C30 is a widely used tool for evaluation of the quality of life in patients with malignant tumors, which consists of 30 questions (most current third version) and has already been used in more than 3 thousand clinical studies across the globe (http://groups.eortc.be/qol/eortc-qlq-c30). Notably, the version of EORTC QLQ-C30 questionnaire translated to polish has already been validated in a clinical setting. The questionnaire provides information on a total of 15 health-related quality of life outcomes including: physical function, role function, cognitive function, emotional function, social function, fatigue, nausea and vomiting, pain, dyspnea, insomnia, appetite loss, constipation, diarrhea, financial impact, and global quality of life. For the purposes of this study, both particular health-related quality of life outcome measures (with particular reference to physical, emotional, and social functions, pain, financial impact, and global quality of life scales) and a summary score derived from the EORTC QLQ-C30 questionnaire will be analyzed. Importantly, robustness of the latter was recently proven by replication and validation of higher order models by Giesinger and coworkers.

In the postoperative period, assessment of health-related quality of life will be performed at the time of discharge from the hospital and at two follow-up visits at 1 and 2 years after the surgical procedure.

Patients will be randomly assigned to one of the two techniques of abdominal closure during the operative procedure and after its main part has been completed, just before the beginning of closure of the abdominal cavity. Randomization will be performed in a 1:1 ratio and in blocks of 20 patients. This procedure of random assignment will be done in the operating room by the nurse through drawing of a sealed envelope with a technique-specific intervention code. Before the beginning of the abdominal closure procedure, a 5 x 5 mm specimen will be procured from the transversalis fascia for further histopathological analyses of the content of collagen in general and specifically, content of type I collagen and type III collagen. Tissue specimens will be fixed in 4% buffered solution of formalin. Subsequently, they will be rinsed in water and automatically processed through 70%, 96% and absolute ethanol solutions, alcohol and xylene solutions, and a series of xylenes. Afterwards, specimens will be embedded in paraffin in a temperature of 60 degrees Celsius to form tissue blocks. The latter will be cut into 3-4 micrometer thick sections using microtome device. Sections will be transferred to microscope slides. First, hematoxylin and eosin staining will be performed and slides will be evaluated for the adequacy of procured tissue sample. For the assessment of overall collagen content, Picrosirius Red staining will be performed and slides will be observed in a microscope coupled with a video camera. Initial evaluation of type I collagen and type III collagen will be assessed by observation of microscopic slides under polarized light. According to the method proposed by Casanova and coworkers, overall collagen content in the obtained specimen will be analyzed by counting the ratio of refringent (collagen-positive) to non-refringent (collagen-negative) areas in 5 randomly selected visual fields observed at 100 magnification. In order to initially establish the content of type I and type III collagen, additional 5 randomly selected visual fields will be assessed at 400 magnification with evaluation of the refringent areas tending to look green, which is characteristic for type III collagen, and those tending to look orange, which is characteristic for type I collagen. In all cases, the final results will be derived from the average of the results from 5 inspected visual fields. Electronic images will be taken and analyzed in a computer image-processing software. A ratio of type I to type III collagen will be calculated.

In order to perform conventional immunohistochemical stainings, specimens will be deparaffinised using xylene, a series of ethanol solutions with decreasing concentrations, and water. The process of antigen retrieval will be performed in a TRIS/EDTA buffer with a pH of 9. Endogenous peroxidase will be blocked with a 3% hydrogen peroxide solution. Protein Block Serum-Free and 2.5% solution of donkey serum will be applied to block non-specific sites of antigen binding. Detection of type I and type III collagen will be performed by using specific anti-collagen I and anti-collagen III primary antibodies, respectively, in an amount of 200 µL per microscopic slide. In the next step, secondary antibodies combined with peroxidase particles will be used. (3-3')diaminobenzidine (1 mL of buffer with 1 drop of substrate) will be used as chromogen, causing brown staining of immune-positive structures. Basophilic structures will be illustrated with hematoxylin contrast staining. Specimens will then be dehydrated with a series of increasing concentration (70%, 96%, and 99.8%) ethanol solutions, acetone, and xylene. Following these procedures, specimens will be closed in an automated fashion. Additionally, double immunofluorescence method will be utilized using secondary antibodies combined with fluorochromes, Vectashield medium, and observation in confocal and fluorescence microscopes.

Closure of the abdominal cavity will be performed with one of the two analyzed surgical techniques, basing on the results of random assignment. Independent of the type of surgical technique, abdominal closure will be done with polydioxanone material (synthetic, monofilament, and slowly absorbable) in the loop form, size 0 according to United States Pharmacopeia, and 36 mm needle. This material is characterized by effective tissue retention period of 90 days and complete absorption period in the range between 182 to 238 days. In both techniques, the intersuture spaces and tissue bites will be of 5 mm, which defines the small bites method. In case of assignment to mass closure technique, abdominal cavity will be closed by a single layer of 2 continuous sutures beginning on the opposite ends of the wound towards the median line and involving peritoneum, transversalis fascia, posterior and anterior layer of rectus abdominis muscle fascia and, in case of incisions beyond the lateral border of rectus abdominis muscle, also the oblique abdominal muscles fascia. In case of assignment to layered closure technique, abdominal cavity will be closed with two separate layers of continuous sutures. The first layer will include peritoneum, transversalis fascia and posterior layer of the rectus abdominis muscle fascia closed with 2 continuous sutures beginning at both ends of the wound and processed medially. In case of incisions not exceeding the lateral border of rectus abdominis muscle, the second layer of 2 continuous sutures will involve only the anterior layer of the rectus abdominis muscle fascia beginning on the lateral ends and processed medially. In case of incisions exceeding the lateral border of rectus abdominis muscle, the second layer of the continuous suture will start at the lateral border of rectus abdominis muscle, process laterally to the lateral end of the wound involving internal oblique abdominal muscle fascia. Subsequently, the suture will be led out on the anterior surface of the external oblique abdominal muscle, put through by the created loop in the medial direction and processed medially to close the external oblique abdominal muscle fascia and anterior layer of the rectus abdominis muscle fascia. Regardless of the applied surgical technique, continuous sutures processing medially will be terminated provided that there is a minimum of 2 cm long segment of overlapping sutures from both sides. Moreover, attention will be given to close only the aponeuroses and to avoid the presence of any muscle or adipose tissue within the sutures. The skin will be closed either with interrupted non-absorbable sutures or with a stapler.

Several parameters will be measured and evaluated intraoperatively. First, the depth of subcutaneous adipose tissue will be measured in centimeters at three sites: median line and bilaterally, in the middle between the median line and lateral end of the wound. Time will be measured from the beginning of abdominal closure to the end of aponeurosis closure. Wound length will be measured in a standard fashion in centimeters. The length of suture used for abdominal closure will be assessed by subtracting the length of the sutures remaining after abdominal closure form the cumulative original length of the sutures. The suture length to wound length ratio will be calculated. Finally, the quality of the abdominal wall with respect to potential risk of burst abdomen and incisional hernia will subjectively be assessed by the operating surgeon in a semiquantitative fashion using the following scale: 1 - poor quality, high risk; 2 - moderate quality, average risk; 3 - good quality, low risk. Inclusion of this variable in risk factor analyses may provide some insight into the reliability of subjective assessment of burst abdomen and incisional hernia by the operating surgeon.

In the immediate postoperative period (until discharge of patient from the hospital) patients will be observed for the development of surgical site infections, defined and classified according to the CDC (Centers for Disease Control) recommendations. More specifically, surgical site infections will be divided into superficial incisional, deep incisional, and organ/space infections. Additionally, each operative procedure will be classified as clean, clean-contaminated, contaminated, and dirty or infected based on the type of operation and intraoperative findings. Data on the duration of postoperative hospitalization (in days) will be collected. All complications of grade III or higher according to Clavien-Dindo classification will be reported in detail. In case of reoperation due to a reason other than occurrence of burst abdomen in the immediate postoperative period, observations for the occurrence of burst abdomen and incisional hernia will be censored.

All patients included in the study will be invited for two follow-up visits at 1 and 2 years (+/- 3 months) following the operative procedure, primarily to evaluate the presence of incisional hernia. During the follow-up visits, a detailed postoperative patient history will be taken, physical examination will be performed, and ultrasonographic examination of the postoperative scar will be done. Regardless of these, images from computed tomography and/or magnetic resonance imaging performed in the course of oncological follow-up protocols (the only additional imaging examination planned as part of this research project is the ultrasonographical examination) will be assessed for the presence of incisional hernia. Data on potential diagnoses and treatment of incisional hernias prior to the follow-up visit in other centers, if such situation occurs, will be collected and analyzed. During each of the follow-up visits, patients will be asked to fill the EORTC QLQ-C30 questionnaires and will undergo skin autofluorescence measurement, as previously described.

Obtained data will be analyzed statistically. Quantitative and qualitative data will be presented as medians with interquartile ranges and numbers with frequencies, respectively. Chi-square test and Fisher's exact test will be used for intergroup comparisons of qualitative variables. Mann-Whitney U test will be applied for intergroup comparisons of quantitative variables. Baseline analysis of the association between the type of surgical technique used for abdominal closure and the occurrence of primary end-point of this study will be performed by comparison of the probability curves for the presence of burst abdomen and incisional hernia over time up to two years with the log-rank test. Analyses of the associations between the remaining factors included in this study and the occurrence of the primary end-point, as well as multivariable analyses will be performed using Cox proportional hazards regression models. Establishment of the optimal cut-offs for quantitative variables for prediction of the development of incisional hernias will be based upon analysis of the receiver operating characteristics curves and Youden index. Initial analysis focused on the risk factors for burst abdomen will be done with Fisher's exact test (surgical technique) and logistic regression models (remaining factors and multivariable analysis). The level of significance was set to 0.05 (two-sided p value). Statistical analyses are planned to be computed in STATISTICA (Dell Inc.) and SAS (SAS Institute) software.

Inclusion Criteria:

• scheduled operative treatment due to malignant tumor of the alimentary system (including liver, bile ducts, and pancreas) through transverse incision in the epigastric region
• provision of informed consent to participate in the study

Exclusion Criteria:

• necessity to perform an urgent operation
• a history of previous surgery performed with transverse incision in the epigastric region
• body mass index >35 kg/m2

Neither the patients included in the study nor the investigators assessing the presence of incisional hernia during follow-up visits will be aware of the type of surgical technique used for abdominal closure (double-blinded study). Only the professionals present in the operating room during surgical procedure, including operator, assistants, anesthesiologist, and nurses, will be aware of the type of assigned intervention.