Methods
Data from potentially transfused patients was collected to a separate "Optimal Use of Blood" (VOK) registry in a joint effort between the Finnish Red Cross Blood Service and ten (out of 21) Finnish hospital districts. Five of the hospital districts were teaching university-connected hospital districts (C, F, G, I and J) and five central hospital districts (A, B, D, E and H) (Fig. 1). The bleeding registry was started in 2002 and continually updated between 2002 and 2011, but the registry has been permanently terminated and erased in 2012. The data collection system is described in more details by Palo and co-workers. Data were collected from hospitals which fulfilled the technical prerequisites and could deliver the required data. Altogether these hospital districts have about 620 000 inpatient episodes annually, constituting 63 % of all Finnish public inpatient hospital episodes. Patient data came from pre-existing electronic medical registers. Computer files provided information on hospital admissions, diagnoses, surgical operations, test results, and blood components as well as on transfusions. The originally collected 90 transfusion-related variables were processed and combined into 149 variables to describe an episode. Systematic audit of data variable was done.
(Enlarge Image)
Figure 1.
Participated Finnish hospital districts
For this study, patients who visited any study hospital district for LC or OC between January 1 2002 and December 31 2007 were extracted from the original VOK registry (2002–2011). In this research registry, data from the hospital district J is only available from Jan 1 2004 to December 31 2007.
ICD-10 (International Statistical Classification of Diseases and Related Health Problems, 10th Revision, 2003) was used to categorize diagnoses. Surgical procedure was defined according to the NCSP (The NOMESCO, Nordic Medico-Statistical Committee, Classification of Surgical Procedures) classification (NCSP–the NOMESCO Classification of Surgical Procedures version 1.7, 2005). All patients over 15 years of age who underwent LC (JKA21) or OC (JKA20) as a primary surgical procedure were included in this study. NCSP does not have a separate procedure code for conversion of LC converted into OC. Thus, conversions cannot be identified from this data and they are reported in the OC group. Cholecystectomies associated with biliary or pancreatic neoplasms were excluded based on the main diagnosis of the operation and diagnoses for the hospital admission. All procedure codes and diagnoses were set by treating physicians.
The following data was collected for each patient: age, sex, American Society of Anaesthesiologists (ASA) class, the main diagnosis, the primary surgical procedure, secondary surgical procedures, the status of the operation (either elective/prescheduled or emergent/acute case), the length of the operation, the usage of blood components during the hospital stay, in-hospital mortality, the length of hospital stay, the hospital, the hospital district and reoperations performed within 60 days of the cholecystectomy (not reported). Emergent operations were performed mostly because of acute cholecystitis or long-lasting painful biliary colic, which required hospital admission. The recorded usage of blood components included the number of transfused red blood cell (RBC) units, number of transfused platelets (PLT), number of transfused fresh frozen plasma products and the total cost of the transfused blood components. The severity of acute cholecystitis (flegmonous/perforated) was not included, because no detailed operation files or pathological reports from removed gallbladders were available.
The Finnish Red Cross Blood Service collects and distributes blood products from non-remunerated donors nationwide in Finland. Each RBC product is equivalent of 1 unit. The PLT product was prepared using a buffy-coat procedure and most of the transfused PLTs were buffy-coat derived. The mean number of PLTs per four-donor product was 296 × 109 and the volume was about 320 ml. Each PLT product is equivalent of (3-)4 units.
The fresh frozen plasma products consisted of fresh frozen plasma (FFP) and Octaplas® (Octapharma Nordic AB). The FFP refers to whole blood–recovered leukodepleted fresh plasma (volume, approximately 270 ml). One product was equal to 1 unit of FFP. During the study period, Octaplas® was available in Finland from 2005 until the end of the study. The usage of the FFP and Octaplas® are presented separately in this study, because the unit size and thus the amount of coagulation factors per unit is smaller in Octaplas® compared to FFP.
The number of all cholecystectomies performed in Finland in 2000–2007 was obtained from the National Institute for Health and Welfare (NIHW) registry.
Permission to use the Bleeding registry data was obtained from the institutional review board of the Finnish Red Cross Blood Service. We did not use patients' identification details (name or social security number) in this register study.
The aims and content of this study are in accordance with the Helsinki Declaration. The respective ethics committee of the Finnish Red Cross Blood Service approved the study protocol.
The data was analyzed using IBM SPSS Statistics 22.0 (IBM, USA 2015). The independence of two categorical variables was tested for with χ test. Mean values of continuous variables were compared with either Student's t-test or Mann–Whitney U test where appropriate. Statistical significance was defined as a p value less than 0.05.