[Can we rely on Norwegian surgery data? A quality control at central and local hospitals of the procedure codes used in the survey on organization of gastrointestinal cancer surgery]
The nation-wide register of hospital discharges in Norway includes ICD-9 and national procedure codes. Hospitals were asked to check five surgical procedures listed in the register against the primary data sources. 649 discharges were controlled. The response rate was 68%. The results indicate that the quality of the data in the register varies for the different procedures. For procedures with high volume (resection of rectum), the error in the register is 3%. This is the same as reported from other Nordic countries. The proportion of errors in the register was high in hospitals with only one registered procedure code. The quality of data can to some extent be checked on the basis of DRG coding (DRG group 468/477). Quality control of register data is required when the number in DRG 468/477 is high in the nation-wide register or when the number of specific procedures in hospitals is low.
Knowledge of how factors related to esophageal cancer resection affect long-term quality of life after surgery is scarce, and no population-based studies are available. Therefore, we conducted a Swedish nationwide, prospective, population-based study of how esophageal surgery-related factors influence quality of life 6 months postoperatively. The Swedish Esophageal and Cardia Cancer register (SECC-register) encompasses 174 hospital departments (97%). Microscopically radically operated patients responded to a validated written questionnaire assessing quality of life. The basic questionnaire (QLQ-C30) and the esophagus-specific module (OES-24) were developed by the European Organization for Research and Treatment of Cancer. The Mann-Whitney test, the Jonckheere-Terpstras test, and logistic regression were used in statistical analyses. Among 100 included patients, the occurrence of surgery-related complications was the main predictor of reduced global quality of life 6 months after surgery (p for trend = 0.03). This effect remained after adjustment for potential confounding variables. Except for anastomotic strictures, each of the predefined complications--i.e., anastomotic leakage, infections, cardiopulmonary complications, and operative technical complications--contributed to decreased quality-of-life scores. Other potentially relevant factors--e.g. degree of lymph node dissection, resection margins, operative blood loss or duration, and hospital type--did not significantly affect quality of life. In conclusion, any measures that can reduce the risk of major surgery-related complications can decrease the negative impact on quality of life after esophageal cancer surgery. More population-based studies are warranted, however.
BACKGROUND: To assess the relationship between hospital volume and early postoperative outcome the incidence and early outcome of all esophagectomies, pancreaticoduodenectomies and gastric resections in Denmark from 1996 to 2004 was described. METHODS: The National Patient Registry and discharge information from all hospital departments were analysed for all the operations due to a malignant diagnosis. All information was examined for postoperative length of stay and hospital mortality. RESULTS: During the study period 26 departments performed at least one esophageal resection, 13 departments performed at least one Whipple procedure and 37 departments performed at least one gastric resection. Four departments performed more than 20 esophageal resections per year, whereas one department performed more than 20 Whipple procedures and one more than 20 gastric resections per year. The overall mean length of stay was 21.6 days, 24 days and 18 days for esophageal, pancreatic and gastric resections, respectively, with no difference between high and low volume departments. The hospital mortality was 8.6%, 8.9% and 8.2%, respectively. CONCLUSION: The overall high mortality and long postoperative stay in patients undergoing upper gastrointestinal cancer surgery in Denmark calls for improvement by regionalisation into 3-4 departments and monitoring of results.
A questionnaire was used to study various aspects of the surgical treatment of cancer of the oesophagus, pancreas, stomach and colon/rectum in Norway. 48 of 51 departments replied. Half of the departments perform radical operations for oesophageal cancer and nearly all do total gastrectomies and low anterior resections for rectal cancer. Palliative surgery is seldom performed for oesophageal cancer, but is frequently performed for gastric cancer. Obstruction of the colon is relieved preferably by primary resection. Low anterior resection is usually performed when a free resection margin of 2 cm can be achieved. Jaundice is relieved palliatively preferably by endoscopy. University hospitals perform total gastrectomy to a greater extent than other hospital and have greater faith in radical operations for pancreatic cancer. Otherwise there are no conspicuous differences of opinion between the different categories of hospitals.
