The Frame Of Statistical Analysis System Of Clinical Data For Gastric Cancer And Study Of Learning Curve Of Laparoscopic D2 Gastrectomy For Cancer

So far, gastric cancer is still the most common malignant tumour in developing countries. It has the highest morbidity after lung cancer, ranking the second in the world. Gastric cancer is also one of the most common malignant tumours in China, its mortality rate stands first among various malignant tumours. There are a great number of clinical cases of gastric cancer in China, which are especially featured by advanced stage. Hospitals at all levels have accumulated an abundance of clinical data sources for a long time, however, problems as following still exist:Firstly,the storage and exchange of clinical data of domestic gastric cancer surgery are at a low information level:looking at countries which have a strong ability of clinical research development such as Japan, America and so on, they all have standards for data-entry nationwide or regionally, simultaneously supported by information products (software, website, etc.). And all of those can satisfy the needs of the nationwide large-scale research, recording and analysis. Up to now, the fact in storage of clinical data in most domestic hospitals is that they fill in the case records, then file and maintain them in the form of medical record, according to the basic requirements of health administrative departments. Since there is no professional standard for data-entry information, the specialized core data is extremely incomplete, and the clinical data mining (CDM) is beset with difficulties. Besides, the data couldn't be exchanged laterally and reciprocally among different hospitals.Secondly, the clinical research in gastric cancer surgery in our country has an insignificant influence in international academic field:in recent years, the surgery technique for gastric cancer has witnessed rapid development, especially the Laparoscopic technique, which is in line with the world-class standard. However, due to a lack of materials that can meet the requirement of international data entry and a effective management of these materials, we have a obvious disadvantage on academic communication with foreign institutions, thus we have few persuasive clinical data. Let alone multi-centers and large sample research.Thirdly, in-depth study should be conducted on the clinical effect of surgical treatment:the current treatment mode of gastric cancer in advanced stage is still comprehensive treatment with surgery as the main method. But there still have a lot remaining to be improved such as the criteria of standard procedure, the controversy of the scope of lymph node dissection needed to be cleared, the relationship between the related factors of surgical treatment and the prognosis, the scientific problems like individualized treatment strategies. There is a dispute about the long-term effect valuation of advancing laparoscopic procedure for gastric cancer and of traditional open surgery. Diagnosis, treatment, prognosis and healthy economic benefits and so on that are relevant to gastric cancer surgery all need in-depth study.The prerequisite to address these problems is the information-based management of these clinical data of gastric cancer surgery so as to meet the basic requirement of a complete, detailed, objective, scientific, general database of international standard. Therefore, we determine to develop independently a whole new statistical analysis system and meet the needs of clinical data mining in gastric cancer surgerySince the first laparoscopic-assisted distal gastrectomy for gastric cancer was reported by Kitano in 1994, and laparoscopy-assisted D2 radical gastrectomy for advanced gastric cancer was reported by Uyama in 1999, various medical treatment units in and abroad are developing in succession laparoscopic surgery for gastric cancer. Meanwhile, they accumulated and summarized some experience. As a new technique, like other laparoscopic surgeries, laparoscopic surgery for gastric cancer certainly has its own special learning curve characteristics. Compared with laparoscopic surgery for colorectal cancer, laparoscopic D2 gastrectomy procedure for gastric cancer is even more difficult and more demanding of skills in terms of technology. Whether it can be performed smoothly depends on the design of surgical procedure, searching of the anatomical plane, and the handling of important blood vessels. Definitely, laparoscopic procedure for gastric cancer is going to be fumbled, improved, enhanced and stabilized during its development in early stage. This paper aims at retrospectively analyzing the law of learning curve of laparoscopic D2 gastrectomy for gastric cancer, which is performed by the same team in our hospital, providing references on how to perform this kind of procedure and how to tide over the learning curve steadily, safely, and quickly for surgeons who have mastered the skill on laparoscopic technic for colorectal cancer proficiently.ChapterⅠThe frame of statistical analysis system of clinical data for gastric cancerObjective:This study aimed to develop a clinical data mining software so as to meet the demand of depth analysis in multicenter clinical researches with large samples within the field of gastric surgeries. The software should also be adapt to international standards as well as clinical habit and at the same time, be capable of conducting both retro-and prospective clinical researches, scientifically depth analyzing and managing the clinical data automatically, and to be more importantly, conducting multilateral cooperations and data-exchange simultaneously.Methods:Basing on the gastric cancer staging system of JGCA-13,14 and UICC staging system, clinical core data including general information, preoperative examination, clinical staging, operation data, surgical staging, pathological staging, final staging, combination therapy, cancer follow-ups, medical expenses, etc. was ascertained. Subsequently, data construction and control logic were both designed in accordance to data mining theory and actual clinical work habits using Microsoft Visual Basic as well as VistaDB programmes. Results:The clinical data of more than 600 patients was collected retrospectively in our hospital during the past year and the data collected were tested to fulfill the designed requirements. To mention that the software, as the database for "Chinese Laparoscopic Gastrointestinal Surgery Study Group--CLASS", was also applied in the first stage of clinical research of "Multicentral retrospective study of the feasibility of laparoscopic procedure for gastric cancer" among 30 hospitals from Chinese mainland and Hong Kong and it could meet the demand for compatibility among these hospitals. The software was commented to be operative, safe, reliable, efficient, and with good compatibility. The highly scientific system was also liked by its automatic classification in cancer staging, etc. In addition, the data mining function could visually realize nested search and complicated analysis and the automatic statistical analysis could be realized in any data for its univariate factor at any time. Inputs and outputs of the records and data could fulfill the demand for further data exchanges and the Output data could be compatible with most of the statistical software.No data error was yet been found during data transmission.Conclusion:The approximate 4,000 indicators contained in the management and analysis system of clinical data for gastric cancer surgery are tested to be comprehensive, scientific, rational, objective, standardized and can effectively provide the foundation for clinical data mining. This gives the clinical researchers a chance for building a standardized platform for gastric cancer surgery by using information techniques and further intensively analyzing and mining the clinical data. After a year's operation, the system is proved to meet the following requirements: computerized management of clinical data; search and statistical analysis of data; provide sufficient and accurate clinical data for the clinical, teaching and scientific research of gastric cancer surgery and related disciplines;supply in the clinical decision and policy making in diagnosis, treatment (operation, chemotherapy, radiotherapy), and prognosis of gastric cancer surgery; provide good foundation for the tracking of medical costs and analysis of health economics; economize in patients' medical costs; Most importantly, standardized data interface makes the large-scale data collection and analysis among hospitals for communication and cooperation possible. The software system has reached its primary designed target. And it is expected to be applied widely in the study of gastric cancer and achieve the goal of extending and upgrading within networks in the future.ChapterⅡStudy of learning curve of laparoscopic D2 gastrectomy for gastric cancerObjective:To analyse the learning curve for one experienced laparoscopic colorectal surgeon converting to laparoscopic D2 gastrectomy for gastric cancer.Methods:From July 2004 to July 2009,a total of 70 patients undergoing laparoscopic D2 gastrectomy by a single surgical team were retrospectively evaluated. Pateints were chronologically divided into 7 groups (group A to G) of 10 cases each, based on surgery date. Patient data including operation time, estimated blood loss (EBL), conversion to open surgery, number of lymph nodes harvest, complications, and recovery indicators(time to flatus, time to liquid diet and length of postoperative hospital stay) were compared among the groups.Results:No statistical differences were found among groups in age, gender, gastrectomy approach, EBL, number of lymph nodes harvest, time to flatus,and length of postoperative hospital stay(p>0.05). In term of operation time, there were no significant differences between group A and B (P=0.999)and among the other 5 groups(C,D,E,F,G)(P>0.05).However, the operation time in group A (300.00±104.59min)and B (261.00±40.50min) were significantly longer than that in the remaining 5 groups respectively(C:191.30±23.11 min, D:188.60±31.38min, E: 181.10±20.18min,F:167.50±32.81min,G:161.30±29.03 min).Compared with group A,time to liquid diet decreased significantly in the remaining 6 groups(P<0.05). In the whole series, there were two conversions to open surgery in group B (convertion rate:2.86%),two intraoperative complications in both group B and C, and one postoperative complication in group C. The morbidity rate was 7.14%.Conclusion:Well-trained laparoscopic colorectal surgeon, by following certain protocol as well as standard surgical procedures, after approximately 20 cases of laparoscopic D2 gastrectomy for gastric cancer, are likely to overcome the learning curve smoothly,safely and swiftly.