Trial Review

The ANZCTR website will be unavailable from 1pm until 3pm (AEDT) on Wednesday the 30th of October for website maintenance. Please be sure to log out of the system in order to avoid any loss of data.

The safety and scientific validity of this study is the responsibility of the study sponsor and investigators. Listing a study does not mean it has been endorsed by the ANZCTR. Before participating in a study, talk to your health care provider and refer to this information for consumers
Trial registered on ANZCTR


Registration number
ACTRN12620000177954
Ethics application status
Approved
Date submitted
11/01/2020
Date registered
17/02/2020
Date last updated
17/02/2020
Date data sharing statement initially provided
17/02/2020
Type of registration
Retrospectively registered

Titles & IDs
Public title
The costs of anaemia after major surgery
Scientific title
The impact of preoperative anaemia on hospital costs following major abdominal surgery
Secondary ID [1] 300235 0
None
Universal Trial Number (UTN)
U111-1246-3429
Trial acronym
Linked study record

Health condition
Health condition(s) or problem(s) studied:
Anaemia 315795 0
Major abdominal surgery 315796 0
Condition category
Condition code
Blood 314084 314084 0 0
Anaemia
Surgery 314085 314085 0 0
Surgical techniques
Anaesthesiology 314086 314086 0 0
Anaesthetics

Intervention/exposure
Study type
Observational
Patient registry
False
Target follow-up duration
Target follow-up type
Description of intervention(s) / exposure
Outcome data from the financial costing records of patients undergoing major abdominal surgery at Austin Health will be collected for the time periods between July 2010 and June 2018. To more clearly appreciate the financial burden associated with preoperative anaemia, we will select four types of subspecialty surgeries, each of which being associated with significant morbidity. These surgeries will include:
• pancreaticoduodenectomy
• major colonic resection
• major rectal resections
• liver resection surgery

We will collect elective and emergent data from the above 4 surgeries using procedural ICD-10 codes
32003-01 Right hemicolectomy with anastomosis
32005-03 Laparoscopic extended right hemicolectomy with anastomosis
32003-03 Laparoscopic right hemicolectomy with anastomosis
32005-01 Extended right hemicolectomy with anastomosis
32004-01 Extended right hemicolectomy with formation of stoma
32000-03 Laparoscopic right hemicolectomy with formation of stoma
32004-03 Laparoscopic extended right hemicolectomy with formation of stoma
32000-01 Right hemicolectomy with formation of stoma
32006-00 Left hemicolectomy with anastomosis
32006-02 Laparoscopic left hemicolectomy with anastomosis
32006-03 Laparoscopic left hemicolectomy with formation of stoma
32006-01 Left hemicolectomy with formation of stoma
32009-00 Total colectomy with ileostomy
32012-01 Laparoscopic total colectomy with ileorectal anastomosis
32009-01 Laparoscopic total colectomy with ileostomy
32012-00 Total colectomy with ileorectal anastomosis
32004-02 Laparoscopic subtotal colectomy with formation of stoma
32005-02 Laparoscopic subtotal colectomy with anastomosis
32005-00 Subtotal colectomy with anastomosis
32004-00 Subtotal colectomy with formation of stoma
32000-02 Laparoscopic limited excision of large intestine with formation of stoma
32003-00 Limited excision of large intestine with anastomosis
32003-02 Laparoscopic limited excision of large intestine with anastomosis
32000-00 Limited excision of large intestine with formation of stoma
30414-00 Excision of lesion of liver
30415-00 Segmental resection of liver
30418-00 Lobectomy of liver
30421-00 Trisegmental resection of liver
30584-00: Pancreaticoduodenectomy with formation of stoma
32024-00 High anterior resection of rectum
32025-00 Low anterior resection of rectum
32026-00 Ultra low anterior resection of rectum
32028-00 Ultra low anterior resection of rectum with hand sutured coloanal anastomosis
32030-00 Rectosigmoidectomy with formation of stoma
32030-01 Laparoscopic rectosigmoidectomy with formation of stoma
32039-00 Abdominoperineal proctectomy
32047-00 Perineal proctectomy
32051-00 Total proctocolectomy with ileo-anal anastomosis
32051-01 Total proctocolectomy with ileo-anal anastomosis and formation of temporary ileostomy
32112-00 Perineal rectosigmoidectomy
92208-00 Anterior resection of rectum, level unspecified

Patients records will reviewed for up for 30 days following discharge from the index admission.
Intervention code [1] 316509 0
Not applicable
Comparator / control treatment
No control group. Comparisons between the groups are not being made. This study is looking at the relationship between anaemia and major abdominal surgery
Control group
Uncontrolled

Outcomes
Primary outcome [1] 322478 0
The primary outcome will be hospital costs of patients who have preoperative anaemia. This will be determined from the hospital medical records
Timepoint [1] 322478 0
Costs will be calculated from day of surgery to hospital discharge. Readmission costs will be considered within 30 days of discharge from the index admission.
Data will be collected from the patients' hospital medical records.
Primary outcome [2] 322599 0
The primary outcome will be hospital costs of patients who do not have preoperative anaemia. This will be determined from the hospital medical records
Timepoint [2] 322599 0
Costs will be calculated from day of surgery to hospital discharge. Readmission costs will be considered within 30 days of discharge from the index admission.
Secondary outcome [1] 378651 0
The development of postoperative complications. Data will be collected from the patients' hospital medical records.
Timepoint [1] 378651 0
These will be collected from day of surgery to hospital discharge. Data will be collected from the patients' hospital medical records.
Secondary outcome [2] 378652 0
The severity of complications as graded by the Clavien Dindo classification. Data will be collected from the patients' hospital medical records.
Timepoint [2] 378652 0
These will be collected from day of surgery to hospital discharge.
Secondary outcome [3] 379779 0
Requirement of allogenic red blood cell (RBC) transfusion and number of RBC transfused. Data will be collected from the patients' hospital medical records.
Timepoint [3] 379779 0
These will be collected from day of surgery to hospital discharge.
Secondary outcome [4] 379780 0
Number of patients with a postoperative ICU admission. Data will be collected from the patients' hospital medical records.
Timepoint [4] 379780 0
These will be collected from completion of surgery to hospital discharge.
Secondary outcome [5] 379781 0
Length of hospital stay. Data will be collected from the patients' hospital medical records.
Timepoint [5] 379781 0
These will be collected from completion of surgery to hospital discharge.
Secondary outcome [6] 379782 0
Readmission within 30 days after discharge. Data will be collected from the patients' hospital medical records.
Timepoint [6] 379782 0
These will be collected from completion of surgery to within 30 days after hospital discharge.
Secondary outcome [7] 379783 0
Costs incurred during readmission. Data will be collected from databases from the hospital's Business Intelligence Unit.
Timepoint [7] 379783 0
These will be collected from the date of readmission to hospital until the date of hospital discharge.

Eligibility
Key inclusion criteria
1. Adult patients (>18 years of age) undergoing one of four types of major abdominal surgeries at Austin Health: Colon resections, Rectal resections, Pancreaticodoudenectomy and Liver resections. Both emergent and elective patients will be included.
2. We will only include patients who underwent surgery of greater than two hours duration and who required at least one overnight hospital stay.
Minimum age
18 Years
Maximum age
No limit
Sex
Both males and females
Can healthy volunteers participate?
No
Key exclusion criteria
Pregnancy

Study design
Purpose
Natural history
Duration
Longitudinal
Selection
Defined population
Timing
Retrospective
Statistical methods / analysis
For statistical analysis, we will use IBM SPSS Statistics for Windows (version 23.0, IBM Corp., Armonk, NY, 2015) and R software 3.5.2 (R Development Core Team, Vienna, Austria, 2018). All collected data will be coded with numerical values, and names of variables will be encrypted to blind the variables’ characteristics to the statistician. Data will be presented as mean ± standard deviation, median (1st ~ 3rd quartiles), or the number of cases (percentile) for descriptive statistics. Estimated values will be described with 95% confidence intervals. Statistical results will be presented with P-values and corresponding effect sizes. Any P-value below 0.05 will considered as statistically significant, based on the null hypothesis significance testing. The number of effects will evaluated with the estimated effect sizes.

We anticipate data from ~1300 patients to be collected.

Prior to statistical analysis, normality will assessed for continuous variables using the Shapiro-Wilk test, a visual check of the normal Q-Q plot, and frequency histograms. For skewed distributions, transformation using logarithmic or exponential scales will be applied, and normality will be re-evaluated using the normal Q-Q plot. If normality is violated post-transformation, non-parametric statistical methods will be applied for that variable. Extreme values will be managed by applying standard statistical methods and then compared to the original values from the data source. If any extreme value cannot be reconciled by interrogating the clinical notes and context of the value, they will be replaced with the Winsorization method.

Variables with a missing rate of greater than 5% will be identified and evaluated to determine whether the missing values occurred at random. In such cases, imputation will be applied using the univariate imputation method of classification and regression trees, using the “mice package”. Maximal iteration will be set at 250, and the number of multiple imputations will be determined appropriately. We will assess the imputed values with an over-imputation diagnostic plot. For the missing analysis of categorical data, the multiple correspondence analysis will be applied with “missMDA package” in the “R” system. Otherwise, statistical analysis will performed as a complete case analysis. Parametric statistical methods will be used only for the continuous, numerical variables that satisfied the normality assumption; otherwise, non-parametric methods will be applied. For parametric, statistical purposes, the homogeneity of variance assumption will be evaluated if required.

For crude estimation of relationship between [Hb]c and hospital costs, curve estimation based on the least square method will be evaluated for linear, quadratic, and cubic models. According to the curve-fit results and physiologically accepted criteria, we will divided [Hb]c into 4 categories; <9.0 g/dL, 9.0 - 13.0 g/dL, 13.0 - 15.0 g/dL, and >15.0 g/dL.

To evaluate the relationship between hospital costs and the patient’s status, correlation analysis and multiple log-linear regression analysis will be used. A stepwise selection method will be applied to determine the relevant regression coefficient for this modelling. For regression, all categorical variables with more than three levels will be transformed into dummy variables. Residuals will be evaluated with the Durbin-Watson statistic test and standardised residual plots, including all partial plots. The variance inflation factor (VIF) will be used to assess collinearity, and Pearson product-moment correlation analysis will be used to evaluate the independency between selected variables. A constant will be included during estimation, in order to compensate for the unknown effect. To enhance the robustness of estimates, bootstrapping will applied to the final regression model, with 1000 samples and the simple sampling method.

To evaluate the effects of [Hb]c on the variables directly related to hospital costs, multivariate general linear modelling will be used. With Pearson’s correlation analysis, the assumption of independence will be validated, and the multivariate normal distribution assumption will be tested using the “mshapiro.test” function in the “mvnormtest package”, in the R system. The Box test will also used for testing homogeneity of covariance matrices. Data for multivariate, general linear modelling will be trimmed using propensity score matching. This will be performed in order to adjust for other variables that may be indicative of patient status, given that they could affect the value of variables directly related to hospital costs. For propensity score matching, the nearest neighbor matching without the caliper method will be applied. Variables selected in the multiple linear regression model will be used for estimating the propensity score. Matched quality will be assessed using standardized differences with a criteria of 0.1, a propensity score distribution graph, and standardized differences density plot check.

Recruitment
Recruitment status
Recruiting
Date of first participant enrolment
Anticipated
Actual
30/01/2020
Date of last participant enrolment
Anticipated
22/02/2020
Actual
Date of last data collection
Anticipated
21/03/2020
Actual
Sample size
Target
1300
Accrual to date
800
Final
Recruitment in Australia
Recruitment state(s)
VIC
Recruitment hospital [1] 15604 0
Austin Health - Austin Hospital - Heidelberg
Recruitment postcode(s) [1] 29003 0
3084 - Heidelberg

Funding & Sponsors
Funding source category [1] 304665 0
Hospital
Name [1] 304665 0
Austin Health
Country [1] 304665 0
Australia
Primary sponsor type
Hospital
Name
Austin Health
Address
Department of Anaesthesia, Austin Health, 145 Studley Road, Heidelberg VIC, 3084
Country
Australia
Secondary sponsor category [1] 304969 0
None
Name [1] 304969 0
Address [1] 304969 0
Country [1] 304969 0

Ethics approval
Ethics application status
Approved
Ethics committee name [1] 305085 0
Austin Health Human Research Ethics Committee
Ethics committee address [1] 305085 0
Ethics committee country [1] 305085 0
Australia
Date submitted for ethics approval [1] 305085 0
13/01/2020
Approval date [1] 305085 0
20/01/2020
Ethics approval number [1] 305085 0
Audit/20/Austin/04

Summary
Brief summary
Trial website
Trial related presentations / publications
Public notes

Contacts
Principal investigator
Name 99218 0
A/Prof Laurence Weinberg
Address 99218 0
Department of Anaesthesia, Austin Health, 145 Studley Road, Heidelberg, VIC 3084
Country 99218 0
Australia
Phone 99218 0
+61 3 94965000
Fax 99218 0
+61 3 94596421
Email 99218 0
[email protected]
Contact person for public queries
Name 99219 0
Laurence Weinberg
Address 99219 0
Department of Anaesthesia, Austin Health, 145 Studley Road, Heidelberg, VIC 3084
Country 99219 0
Australia
Phone 99219 0
+61 3 94965000
Fax 99219 0
+61 3 94596421
Email 99219 0
[email protected]
Contact person for scientific queries
Name 99220 0
Laurence Weinberg
Address 99220 0
Department of Anaesthesia, Austin Health, 145 Studley Road, Heidelberg, VIC 3084
Country 99220 0
Australia
Phone 99220 0
+61 3 94965000
Fax 99220 0
+61 3 94596421
Email 99220 0
[email protected]

Data sharing statement
Will individual participant data (IPD) for this trial be available (including data dictionaries)?
No
No/undecided IPD sharing reason/comment
As this is an observational study, patients have not provided informed consent for their data to be shared.


What supporting documents are/will be available?

Doc. No.TypeCitationLinkEmailOther DetailsAttachment
6381Statistical analysis plan  [email protected] 379040-(Uploaded-11-01-2020-14-54-08)-Study-related document.pdf
6829Ethical approval  [email protected] 379040-(Uploaded-09-02-2020-19-54-58)-Study-related document.pdf



Results publications and other study-related documents

Documents added manually
No documents have been uploaded by study researchers.

Documents added automatically
SourceTitleYear of PublicationDOI
EmbaseAssociations between preoperative anaemia and hospital costs following major abdominal surgery: Cohort study.2021https://dx.doi.org/10.1093/bjsopen/zraa070
N.B. These documents automatically identified may not have been verified by the study sponsor.