Abstract
Background
Clinicians are increasingly confronted with elderly patients across nearly all surgical disciplines. Advanced age is an independent risk factor for complications, prolonged hospitalization, and particularly for red blood cell (RBC) transfusion. The elderly suffer more frequently from anemia, which is associated with higher perioperative transfusion requirements, increased complication rates, and mortality Surgical disciplines differ not only in primary diagnoses but also in the demographics of their typical patient population.
Material and methods
Hospitalized surgical patients in Germany between January 1st, 2013, and December 31st 2022, were analyzed. Inclusion criteria: Age ≥65 years, surgical disciplines: General surgery, Gynecology, Cardiac surgery, Thoracic surgery, Oral and Maxillofacial surgery, Neurosurgery, Trauma and Orthopedic surgery, Urology or Vascular surgery. Exclusion criteria included unknown or overlapping surgical procedures. Risk factors, comorbidities, transfusion rates, bleeding, complications, and blood product use were assessed using ICD and OPS (the German national procedure coding system) codes. The primary endpoint was RBC transfusion rate; the secondary endpoint was preoperative transfusion rate.
Results
A total of 21,143,317 patients were analyzed of which 2,640,608 (12%) received RBC transfusion. The highest transfusion rate was seen in Trauma and Orthopedic surgery (17.2% - No.=1,116,082). Trauma and Orthopedic surgery also had the highest risk (Odds Ratio [OR]) for RBC Transfusion (1.36 [1.35–1.36]). The highest preoperative transfusion rate was noted in general surgery, where 48.2% (No.=384,553) of the transfused patients received at least one RBC transfusion prior to surgery.
Discussion
Our study revealed that the highest transfusion rate in elderly patients is in Trauma and Orthopedic surgery, while the highest rate of preoperative transfusion was seen in General surgery. It would seem worthwhile to reduce these rates wherever possible, and these are the disciplines which offer the most room for improvement and would thus benefit most from consistent establishment of PBM programs.
Keywords: blood transfusion, elderly, secondary data analyses, nationwide, Patient Blood Management (PBM)
INTRODUCTION
With life expectancy increasing worldwide, clinicians are facing an ever-greater number of elderly patients1. This results in a higher prevalence of comorbidities, reduced physiological reserves, and increased complications2–4. There is no universally accepted definition of “elderly” patients, since chronological age does not reflect the wide variation between a healthy individual and one with multimorbidity. The term commonly used to reflect the multidimensionality of “chronological” age, is frailty, which captures functional and physiological deficits in multiple systems. Vermeiren et al. described frailty as a prediction for negative health outcomes with a 1.8- to 2.3-fold risk for mortality5. The German Society of Geriatrics guideline ‘Comprehensive Geriatric Assessment (CGA) for hospitalized patients’ defines a “target population” of patients of 65 years and older6. A study by Blum et al. demonstrated that increasing age is an independent risk factor for complications, longer hospital stays and in particular, the need for RBC transfusion1. In general, elderly suffer more frequently from anemia2,3, which is also known as a preoperative predictor for perioperative red blood cell (RBC) transfusion, higher complication rates and increased mortality4,7–9. In elderly and potentially frail surgical patients, the optimal transfusion strategy is a matter of ongoing debate10–12.
Surgical disciplines differ not only in the type of procedure associated with significant blood loss, but also in the phenotype of their typical patients, for example, in terms of pre-existing conditions and demographics. For this reason, it is vital to differentiate between surgical disciplines. With this in mind, the present study was designed to provide an overview of blood transfusion behavior in elderly, surgical patients across different surgical disciplines. Due to the large cohort, our results may provide a basis for further studies, so as to offer more tailored transfusion practices within the surgical disciplines. In order to test our hypothesis that transfusion rates vary across the major surgical disciplines, we compared intraoperative and preoperative RBC transfusion rates in these disciplines among patients ≥65 years undergoing surgery: the primary endpoint was the intraoperative RBC transfusion rate among surgical disciplines, while the secondary endpoint was the preoperative RBC transfusion rate.
MATERIAL AND METHODS
Availability of data
In Germany, hospitals are legally required to report relevant diagnoses using the International Statistical Classification of Diseases and Related Health Problems (ICD) codes and the International Statistical Classification of Procedures (OPS) codes13 (Supplementary Content, Table SI). According to §21 of the German Hospital Finance Law (KHG), hospitals must submit reimbursement data to the Institute for Hospital Remuneration (InEK) for development of the DRG system. These data are forwarded to the Federal Statistical Office (DESTATIS), anonymized, and made available to researchers under a data usage contract. Reporting of ICD-10 and OPS codes, reinforced by reimbursement incentives and monitored by health insurance medical services, ensures high completeness and accuracy. Data are stored locally by DESTATIS, and no patient or hospital identifiers were accessible to the authors. As the dataset was institutionally anonymized, General Data Protection Regulation (GDPR) does not apply. The Ethics Committee of the University Hospital Frankfurt waived the requirement for ethical approval and informed consent (Chair: Prof. Dr. Harder, Ref: 2022-766). Data collection was structured, representative, and conducted in accordance with the Declaration of Helsinki and STROBE guidelines.
Inclusion criteria
Because there is no standardized definition of “elderly”, we chose 65 years as the cut-off age, which is commonly used in the literature and is in line with geriatric guidelines6. Therefore, 65 years seemed a reasonable cutoff value for the purpose of our analysis. Hospitalized surgical patients ≥65 years in Germany between January 1st, 2013, and December 31st, 2022, treated in General surgery, Gynecology, Cardiac surgery, Thoracic surgery, Oral and Maxillofacial surgery, Neurosurgery, Trauma and Orthopedic surgery, Urology or Vascular surgery were included in the analysis (Figure 1).
Figure 1.
Flowchart showing the screening and exclusion criteria in hospitalized surgical patients in Germany (2013–2022)
RBC: red blood cell.
Exclusion criteria
Patients treated by multiple surgical disciplines or unknown surgical disciplines were excluded. Furthermore, patients were excluded if the coded procedure did not match with one of the disciplines analyzed (General surgery, Gynecology, Cardiac surgery, Thoracic surgery, Oral and Maxillofacial surgery, Neurosurgery, Trauma and Orthopedic surgery, Urology and Vascular surgery) (Figure 1).
Definitions and data acquisition
The variables analyzed (Supplementary Content, Table SI) were assessed in relation to transfusion status (transfused vs non-transfused) and surgical discipline. Diagnoses were coded using ICD-10-GM, and procedures classified according to the OPS catalog, with details provided in Supplementary Content, Table SI.
Patients were divided into transfused and non-transfused groups and stratified by surgical discipline. We examined RBC transfusion rates, RBC count, and timing (pre-, intra-, or post-operative). Administration of fibrinogen, prothrombin complex concentrate (PCC), platelets, and fresh frozen plasma (FFP) were also analyzed.
RBC unit counts were grouped as follows: 1–5, 6–10, 11–15, 16–23, 24–31, 32–39, 40–47, 48–55, 56–63, 64–71, and 72–79, defined by ICD and OPS codes. Multimorbidity was assessed using the Elixhauser Comorbidity Index (ECI), which incorporates 31 comorbidities. The ECI as a measure of overall severity of comorbidities uses ICD-based comorbidity measures. It is able to outperform other scores such as the Charleston Comorbidity Index (CCI) in predicting in-hospital outcomes, including in-hospital mortality, length of stay, and hospital resource utilization14. Perioperative transfusion represents an acute, in-hospital outcome rather than a long-term prognosis, while the Elixhauser index provides an appropriate and sensitive measure of comorbidity-associated risk in the context of our study. According to the Federal Statistical Office, massive transfusion was defined as >5 RBC units per patient, a predefined definition13. To mark this difference, the term “massive transfusion” as defined above is referred to in the manuscript as “massive transfusion (Destatis [Des.])”. In addition to cell salvage as a key blood-conservation strategy, we analyzed blood products including prothrombin complex concentrate (PCC) and fresh frozen plasma (FFP). In the course of the study period, new transfusion guidelines were introduced (2020), defining individualized Hb thresholds: hemorrhagic shock <7–8 g/dL; Cardiac surgery without acute bleeding <7.5 g/dL; trauma/orthopedic patients >65 years or with severe cardiovascular disease <8 g/dL; stroke, subarachnoid hemorrhage, or traumatic brain injury <8 g/dL; intensive care patients without acute bleeding <7 g/dL15. In older guidelines, an Hb value <6 g/dL was considered the transfusion threshold. In the range of 6–8 g/dL, transfusion is recommended in cases of limited compensation, risk factors such as underlying cardiovascular disease, or presence of signs of anemic hypoxia (such as tachycardia, hypotension and lactic acidosis)16.
Statistical analysis
Categorical variables are presented as absolute frequencies and percentages. Continuous variables (age, length of stay [LOS] and mechanical ventilation hours) were assessed for normality and found to be non-normally distributed. They are therefore reported as medians with interquartile ranges (IQR; 25th–75th percentile).
Group comparisons were considered statistically significant due to the large sample sizes. Predefined variables including comorbidities, RBC transfusions, anemia, bleeding, complications, and use of blood products were analyzed based on ICD and OPS coding (Supplementary Content, Table SI). Multiple logistic regression models were constructed to estimate the odds of RBC transfusion by surgical discipline.
All data preprocessing was conducted using Microsoft Excel 2019 (Microsoft Corp., Redmond, WA, USA), and statistical analyses were performed using SAS software, version 9.4M6 (SAS Institute Inc., Cary, NC, USA).
The primary endpoint was intraoperative RBC transfusion rate among surgical disciplines; secondary endpoint was preoperative RBC transfusion rate.
RESULTS
Between January 1st, 2013, and December 31st, 2022, No.=82,262,656 patients (≥65 years) underwent surgery, of which No.=60,119,339 had to be excluded because of overlapping surgical disciplines (No.=667,766 transfused and No.=1,853,848 non-transfused) or unknown/excluded surgical disciplines (No.=2,492,293 transfused and v56,105,432 non-transfused).
In total, No.=21,143,317 patients were analyzed, of which No.=2,640,608 (12%) received RBC transfusion (Figure 1, Table I). The highest chance for RBC Transfusion was seen in Trauma and Orthopedic surgery (OR 1.36 [1.35–1.36]) (Supplementary Content, Table SIV). The highest preoperative transfusion rate was seen in General surgery. Of the patients transfused in General surgery, 48.2% (No.=384,553) received at least one RBC transfusion prior to surgery (Table II).
Table I.
Demographic of all surgical patients ≥ 65 years among surgical disciplines divided into No-RBC and RBC transfusion
| General surgery | Gynecology | Cardiac surgery | Thoracic surgery | Oral and Maxillofacial surgery | Neurosurgery | Trauma and Orthopedic surgery | Urology | Vascular surgery | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| No. | % | No. | % | No. | % | No. | % | No. | % | No. | % | No. | % | No. | % | No. | % | ||
| Total patients | No RBC | 4,973,925 | 86.2 | 859,813 | 98.7 | 1,499,166 | 83.1 | 291,400 | 83.8 | 97,896 | 97.2 | 419,314 | 92.3 | 6,472,806 | 83 | 2,496,043 | 94.2 | 1,392,346 | 83.2 |
| RBC | 798,080 | 13.8 | 10,904 | 1.3 | 253,531 | 16.9 | 47,288 | 16.2 | 2,733 | 2.8 | 32,421 | 7.7 | 1,116,082 | 17.2 | 145,311 | 5.8 | 234,258 | 16.8 | |
| Female | No RBC | 2,249,346 | 45.2 | 851,205 | 99 | 588,959 | 39.3 | 114,142 | 39.2 | 50,686 | 51.8 | 207,974 | 49.6 | 4,151,561 | 64.1 | 497,487 | 19.9 | 599,552 | 43.1 |
| RBC | 376,616 | 47.9 | 10,803 | 99.1 | 107,545 | 42.4 | 16,783 | 35.5 | 1,274 | 46.6 | 16,306 | 50.3 | 782,916 | 70.2 | 31,062 | 21.4 | 110,071 | 47 | |
| ICU | No RBC | 269,276 | 5.4 | 6,341 | 0.7 | 151,581 | 10.1 | 54,160 | 18.6 | 3,577 | 3.7 | 73,967 | 17.6 | 199,168 | 3.1 | 51,348 | 2.1 | 113,338 | 8.1 |
| RBC | 158,549 | 19.9 | 1,408 | 12.9 | 81,534 | 32.2 | 16,360 | 34.6 | 861 | 31.5 | 13,265 | 40.9 | 180,064 | 16.1 | 21,641 | 14.9 | 47,245 | 20.2 | |
| Age; years, median (Q1; Q3) | No RBC | 76 (70;81) | 74 (69;79) | 78 (73;83) | 74 (69;78) | 76 (71;82) | 75 (70;80) | 75 (70;81) | 75 (70;80) | 75 (70;80) | |||||||||
| RBC | 78 (73;84) | 78 (73;83) | 76 (71;80) | 75 (70;88) | 78 (72;84) | 76 (71;81) | 82 (76;87) | 78 (73;84) | 76 (71;81) | ||||||||||
| ECI; No., median (Q1; Q3) | No RBC | 5 (0;10) | 0 (0;5) | 9 (5;15) | 7 (3;15) | 0 (0;6) | 5 (0;11) | 0 (0;5) | 0 (0;5) | 5 (0;12) | |||||||||
| RBC | 13 (6;20) | 10 (3;17) | 13 (7;19) | 14 (8;21) | 10 (5;17) | 14 (8;21) | 7 (2;14) | 10 (5;17) | 12 (6;19) | ||||||||||
| Length of stay; hours, median (Q1; Q3) | No RBC | 123 (53;243) | 103 (73;159) | 153 (66;270) | 239 (150;382) | 122 (72;187) | 184 (79;337) | 195 (102;291) | 101 (67;173) | 148 (68;269) | |||||||||
| RBC | 354 (188;630) | 311 (198;488) | 340 (245;522) | 495 (311;770) | 338 (207;561) | 427 (232;698) | 365 (261;588) | 311 (194;497) | 410 (247;673) | ||||||||||
| Ventilation; hours, median (Q1; Q3) | No RBC | 44 (11;168) | 11 (4;45) | 13 (5;41) | 23 (5;84) | 30 (14;73) | 85 (28;225) | 18 (4;52) | 28 (7;97) | 25 (6;69) | |||||||||
| RBC | 120 (31;388) | 40 (9;161) | 35 (12;120) | 117 (27;386) | 74 (29;218) | 197 (68;387) | 41 (11;144) | 55 (14;212) | 53 (16;192) | ||||||||||
RBC: red blood cell; ECI: Elixhauser comorbidity index.
Table II.
Use of blood products among surgical disciplines
| General surgery | Gynecology | Cardiac surgery | Thoracic surgery | Oral and Maxillofacial surgery | Neurosurgery | Trauma and Orthopedic surgery | Urology | Vascular surgery | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| RBC administration prior to surgery | ||||||||||||||||||
| Not stated | 444 | 0.06 | 0 | 0 | 74 | 0.03 | 0 | 0 | 0 | 0 | 0 | 0 | 50 | 0 | 15 | 0.01 | 36 | 0.02 |
| No | 413,083 | 51.76 | 7,501 | 68.79 | 175,790 | 69.34 | 31,755 | 67.15 | 1,780 | 65.13 | 24,620 | 75.94 | 907,058 | 81.27 | 90,688 | 62.41 | 147,574 | 63.00 |
| Yes | 384,553 | 48.18 | 3,403 | 31.21 | 77,667 | 30.64 | 15,533 | 32.85 | 953 | 34.87 | 7,801 | 24.06 | 208,974 | 18.72 | 54,608 | 37.58 | 86,648 | 36.99 |
| Timepoint of RBC administration | ||||||||||||||||||
| Admission to surgery; hours, median (Q1;Q3) | 50 (18;160) | 27 (20;118) | 51 (24;142) | 72 (24;215) | 38 (17;134) | 8 (1;66) | 23 (12;50) | 30 (18;119) | 89 (23;240) | |||||||||
| Admission to RBC; hours, median (Q1;Q3) | 43 (6;172) | 66 (24;165) | 71 (25;156) | 102 (28;267) | 57 (17;169) | 55 (5;209) | 66 (26;144) | 54 (19;150) | 84 (24;217) | |||||||||
| Surgery to RBC; hours, median (Q1;Q3) | 0.05 (−47;45) | 20 (−14;73) | 2.45 (−1.6;24.9) | 5.10 (−11;72) | 6 (−19;59) | 9.55 (0;113) | 24 (1;75) | 5 (−22;−16) | 3.8 (−29;49) | |||||||||
| RBC prior to surgery; hours, median (Q1;Q3) | −49 (−139;−17) | −42 (−145;−18) | −18 (−71;−2) | −50 (−188;−13) | −46 (−142;−16) | −7(−67;−1) | −19 (−50;−3) | −44 (−121;−16) | −101 (−266;−21) | |||||||||
| RBC after surgery; hours, median (Q1;Q3) | 41 (5;143) | 48 (16;106) | 7 (2;49) | 37 (4;140) | 38 (6;109) | 43 (3;178) | 40 (9;97) | 45 (8;115) | 28 (5;99) | |||||||||
| Blood products | ||||||||||||||||||
| No. | % | No. | % | No. | % | No. | % | No. | % | No. | % | No. | % | No. | % | No. | % | |
| RBCs | 798,080 | 100 | 10,904 | 100 | 253,531 | 100 | 47,288 | 100 | 2,733 | 100 | 32,421 | 100 | 1,116,082 | 100 | 145,311 | 100 | 234,258 | 100 |
| Platelets | 23,403 | 2.93 | 144 | 1.32 | 43,213 | 17.04 | 2,388 | 5.05 | 147 | 5.38 | 5,320 | 16.41 | 11,233 | 1.01 | 3,207 | 2.21 | 17,509 | 7.47 |
| Fresh Frozen Plasma | 86,238 | 10.81 | 531 | 4.87 | 45,890 | 18.10 | 6,542 | 13.83 | 205 | 7.50 | 5,084 | 15.68 | 58,384 | 5.23 | 8,367 | 5.76 | 22,434 | 9.58 |
| Prothrombin complex concentrate | 59,584 | 7.47 | 200 | 1.83 | 35,771 | 14.11 | 3,663 | 7.75 | 129 | 4.72 | 7,112 | 21.94 | 30,103 | 2.70 | 4,139 | 2.85 | 11,690 | 4.99 |
| Fibrinogen | 21,237 | 2.66 | 183 | 1.68 | 39,489 | 15.58 | 2,182 | 4.61 | 73 | 2.67 | 3,773 | 11.64 | 15,424 | 1.38 | 2,672 | 1.84 | 10,977 | 4.69 |
| Cell salvage | 963 | 0.12 | * | * | 23,812 | 9.39 | 166 | 0.35 | * | * | * | * | 56,015 | 5.02 | 124 | 0.09 | 9,353 | 3.99 |
| Massive blood transfusion (des.) | 30,874 | 3.87 | 93 | 0.85 | 10,274 | 4.05 | 2,671 | 5.65 | 41 | 1.50 | 653 | 2.01 | 19,183 | 1.72 | 3,045 | 2.10 | 9,171 | 3.91 |
Characteristics by surgical discipline
Most of the patients over 65 who were analyzed were treated by Trauma and Orthopedic surgery, followed by General surgery. The least common discipline was Oral and Maxillofacial. The highest median age was seen in transfused patients in Trauma and Orthopedic surgery and in non-transfused patients in Cardiac surgery (78 [73;83]). When divided into age groups of 5 years, Gynecology was the most common discipline among those aged 65–69, followed by Thoracic surgery in patients from 70 to 74 years. From 75 up to 89 years Cardiac surgery was the most common discipline. From 90–110 years Trauma and Orthopedic surgery was the most common (Table III). Besides Gynecology, the proportion of female patients was the highest in Trauma and Orthopedic surgery. The highest Intensive Care Unit (ICU) admission rate was seen in Thoracic surgery (21% - No.=70,520), while the lowest was in Gynecology (1% - No.=7,749). The median Elixhauser-Score was significantly higher in transfused patients. Among transfused patients the highest median Elixhauser-Score was observed in Thoracic- and Neurosurgery. The highest Intensive Care Unit (ICU) admission rate in transfused patients was seen in Neurosurgery (40.91%), while the highest ICU-admission rate in non-transfused patients was seen in Thoracic surgery (18.6%). ICU admission was significantly more frequent in transfused patients: it was 3.6 times higher in General surgery and up to 17.4 times higher in Gynecology (Table I). The most common risk factor was essential hypertension with a rate of 62.2%. In transfused patients, this risk factor was present most frequently in Trauma and Orthopedic surgery, with a rate of 62% (No.=863,181) and in non-transfused patients in Vascular surgery with 60% (No.=863,181). Acute kidney injury (AKI) was the most frequently observed complication. Irrespective of the transfusion group, Vascular surgery was the leading. (Table IV and Supplementary Content, Table SII)
Table III.
Heat map of Red Blood Cell-transfusion rate in % among surgical disciplines divided by age groups green =low red= high
| Age (y) | General surgery | Gynecology | Cardiac surgery | Thoracic surgery | Oral and Maxillofacial surgery | Neurosurgery | Trauma and Orthopedic surgery | Urology | Vascular surgery |
|---|---|---|---|---|---|---|---|---|---|
| 65–69 | 9.37 | 0.67 | 19.6 | 3.29 | 1.95 | 6.44 | 5.79 | 3.45 | 11.68 |
| 70–74 | 11.25 | 0.87 | 18.7 | 4.46 | 2.35 | 6.56 | 8.32 | 4.25 | 12.98 |
| 75–79 | 13.60 | 1.27 | 16.4 | 7.20 | 2.69 | 7.05 | 12.45 | 5.21 | 14.43 |
| 80–84 | 16.84 | 1.91 | 11.5 | 14.21 | 3.04 | 8.03 | 19.17 | 6.93 | 16.80 |
| 85–89 | 20.05 | 3.20 | 8.0 | 28.61 | 3.71 | 9.23 | 28.67 | 9.41 | 20.11 |
| 90–94 | 21.92 | 4.57 | 5.9 | 43.63 | 4.04 | 8.75 | 36.88 | 11.81 | 22.39 |
| 95–110 | 21.80 | 4.54 | 3.8 | 54.77 | 4.02 | 6.80 | 42.77 | 14.16 | 20.16 |
Table IV.
Risk factors, complications and mortality in all surgical patients ≥ 65 years across surgical disciplines divided by RBC- and No-RBC
| General surgery | Gynecology | Cardiac surgery | Thoracic surgery | Oral and Maxillofacial surgery | Neurosurgery | Trauma and Orthopedic surgery | Urology | Vascular surgery | ||
|---|---|---|---|---|---|---|---|---|---|---|
| % | % | % | % | % | % | % | % | % | ||
| Total patients | No RBC | 86.2 | 98.7 | 83.1 | 83.8 | 97.2 | 92.3 | 83.0 | 94.2 | 83.2 |
| RBC | 13.8 | 1.3 | 16.9 | 16.2 | 2.8 | 7.7 | 17.2 | 5.8 | 16.9 | |
| Risk factors | ||||||||||
| Essential hypertension | No RBC | 58.5 | 53.1 | 61.1 | 55.9 | 51.1 | 61.2 | 61.6 | 59.6 | 62.0 |
| RBC | 57.4 | 62.1 | 63.5 | 55.3 | 58.1 | 60.5 | 65.9 | 59.6 | 60.6 | |
| Coronary artery disease | No RBC | 15.6 | 5.3 | 46.2 | 18.5 | 12.3 | 12.9 | 11.4 | 16.2 | 24.3 |
| RBC | 26.0 | 14.0 | 67.0 | 67.0 | 22.0 | 18.0 | 20.0 | 25.0 | 33.0 | |
| Anticoagulation therapy | No RBC | 14.5 | 8.3 | 29.7 | 13.7 | 19.7 | 12.8 | 12.9 | 13.6 | 17.6 |
| RBC | 20.6 | 21.1 | 18.8 | 18.5 | 22.5 | 13.2 | 19.8 | 19.1 | 21.3 | |
| Congestive heart failure | No RBC | 10.5 | 4.0 | 44.5 | 12.9 | 6.9 | 6.7 | 8.2 | 6.8 | 14.8 |
| RBC | 29.4 | 20.3 | 58.0 | 29.1 | 19.9 | 15.8 | 24.0 | 19.8 | 30.0 | |
| Cardiac arrhythmias | No RBC | 21.9 | 10.6 | 66.0 | 23.3 | 18.0 | 20.2 | 16.4 | 17.3 | 24.7 |
| RBC | 42.2 | 32.0 | 64.4 | 44.7 | 35.0 | 36.6 | 32.7 | 32.8 | 36.6 | |
| Chronic pulmonary disease | No RBC | 9.3 | 5.4 | 20.0 | 27.6 | 6.9 | 7.5 | 8.0 | 7.6 | 11.7 |
| RBC | 13.6 | 8.5 | 13.9 | 27.6 | 11.3 | 8.7 | 10.4 | 10.1 | 15.5 | |
| Metastatic cancer | No RBC | 4.9 | 10.4 | 0.2 | 20.3 | 4.6 | 3.2 | 0.4 | 2.5 | 8.9 |
| RBC | 10.6 | 25.5 | 0.3 | 17.7 | 13.4 | 6.5 | 1.3 | 12.1 | 9.1 | |
| Solid tumor without metastasis | No RBC | 5.5 | 5.4 | 1.0 | 26.4 | 11.2 | 7.2 | 1.2 | 9.7 | 4.7 |
| RBC | 10.1 | 12.9 | 1.6 | 16.9 | 19.1 | 10.2 | 2.8 | 21.7 | 6.4 | |
| Alcohol abuse | No RBC | 0.4 | 0.0 | 0.2 | 0.3 | 0.5 | 0.4 | 0.3 | 0.1 | 0.2 |
| RBC | 3.5 | 0.8 | 1.0 | 2.5 | 3.5 | 3.5 | 1.6 | 1.1 | 1.6 | |
| Obesity | No RBC | 6.4 | 7.5 | 8.1 | 5.8 | 3.1 | 6.0 | 10.7 | 7.0 | 10.5 |
| RBC | 5.8 | 11.2 | 12.7 | 5.1 | 3.0 | 3.7 | 7.6 | 4.8 | 6.0 | |
| Complications | ||||||||||
| Stroke | No RBC | 0.3 | 0.1 | 0.7 | 0.3 | 0.2 | 1.8 | 0.2 | 0.1 | 1.1 |
| RBC | 1.2 | 0.7 | 3.1 | 1.3 | 1.3 | 6.7 | 1.0 | 0.7 | 2.1 | |
| CPR | No RBC | 0.4 | 0.1 | 1.6 | 0.8 | 0.2 | 0.6 | 0.3 | 0.1 | 0.5 |
| RBC | 3.5 | 1.1 | 6.4 | 7.0 | 4.2 | 5.2 | 1.6 | 1.9 | 4.0 | |
| Pneumonia | No RBC | 3.0 | 0.2 | 3.3 | 9.6 | 0.7 | 4.9 | 1.3 | 0.8 | 2.6 |
| RBC | 15.8 | 5.1 | 12.9 | 32.7 | 14.2 | 31.3 | 8.4 | 7.7 | 14.7 | |
| AKI | No RBC | 14.9 | 5.2 | 29.4 | 15.2 | 9.7 | 9.4 | 12.3 | 14.2 | 30.4 |
| RBC | 33.4 | 24.7 | 33.6 | 28.5 | 24.8 | 18.4 | 31.4 | 34.7 | 44.0 | |
| RRT | No RBC | 0.6 | 0.7 | 0.8 | 0.6 | 0.4 | 0.4 | 0.3 | 0.4 | 10.7 |
| RBC | 6.1 | 1.6 | 10.7 | 8.6 | 3.9 | 5.2 | 2.0 | 4.7 | 20.3 | |
| Complication after transfusion | No RBC | |||||||||
| RBC | 1.4 | 1.5 | 0.6 | 0.9 | 0.9 | 0.9 | 0.8 | 0.9 | 1.7 | |
| Pulmonary embolism | No RBC | 0.3 | 0.1 | 0.2 | 0.8 | 0.1 | 0.6 | 0.2 | 0.1 | 0.4 |
| RBC | 1.3 | 1.6 | 0.4 | 2.3 | 1.4 | 2.4 | 0.9 | 1.1 | 1.0 | |
| Myocardial infarction | No RBC | 0.4 | 0.1 | 1.3 | 0.4 | 0.2 | 0.3 | 0.3 | 0.2 | 0.6 |
| RBC | 2.2 | 1.3 | 4.1 | 2.0 | 2.6 | 2.2 | 1.7 | 2.0 | 3.4 | |
| Mortality | No RBC | 1.5 | 0.1 | 0.7 | 3.8 | 0.5 | 5.4 | 1.1 | 0.6 | 2.3 |
| RBC | 15.3 | 4.0 | 6.9 | 21.8 | 9.9 | 34.5 | 7.2 | 7.5 | 12.6 | |
Heat map green = low red = high for each risk-factor and comorbidity.
RBC: red blood cell; CPR: cardiopulmonary resuscitation; AKI: acute kidney injury; RRT: renal replacement therapy.
Transfused patients
Trauma and Orthopedic surgery showed the highest RBC transfusion rate, at 17.2% (No.=1,116,082), while Gynecology had the lowest rate at 1.3% (No.=10,904). The most numerous age group among transfused patients in Trauma and Orthopedic surgery were those aged 80–84 (24% - No.=267,993). The youngest group, aged 65–69, had the biggest share in Thoracic surgery: the detailed age groups among the transfused are listed in Supplementary Content, Figures S1 and S2. Across all surgical disciplines, transfused patients exhibited an age-related shift, with the proportion of RBC-transfused patients exceeding that of non-transfused patients at distinct age thresholds. Cardiac surgery showed this at age 65–69, whereas in Gynecology, Thoracic surgery and Vascular surgery, the shift was seen at age 75–79. In General surgery, Oral and Maxillofacial surgery, Neurosurgery, Trauma and Orthopedic surgery, and Urology, the shift came at age 80–84. Detailed age-stratified distributions for each discipline are presented in Figure 2 A-I. The highest transfusion rate prior to surgery was observed in General surgery with 48.2% (No.=384,553), while the lowest was in Trauma and Orthopedic surgery with 18.7% (No.=208,974).
Figure 2.
A-I - Displaying the share of transfused and non-transfused patients among age groups in every surgical discipline
The longest median timespan between admission and surgery among transfused patients was seen in Vascular surgery (89 hours [h] [23 h; 240 h]). However, the longest median timespan between admission and the first RBC transfusion was seen in Thoracic surgery (102 h [28 h; 267 h]). The highest platelet transfusion rate among RBC-transfused patients was seen in Cardiac surgery (17% - No.=43,213), followed by Neurosurgery (16.4% - No.=5,320). In every discipline, most transfusions were in the order of 1–5 RBC per patient. With regard to 6–10 RBC per patient, Cardiac surgery had the highest share with 12.9% (No.=32,726) (Supplementary Content, Table SIV). Looking at transfusion rates over the years, the steepest decline was seen in Cardiac surgery from 18.4% (No.=31,732) in 2013 down to 11.9% (No.=19,562) in 2022 (Supplementary Content, Figure S2).
In a logistic regression to determine the chance of an RBC transfusion an odds ratio (OR) over 1 was seen in General surgery, Cardiac surgery, Thoracic surgery, Trauma and Orthopedic surgery and Vascular surgery, with the highest OR in Trauma and Orthopedic surgery (1.36 [1.35–1.36]). For detailed numbers see Supplementary Content, Table IV. The final model was derived relative to General surgery, OR over 1 was seen in Cardiac, Thoracic, Trauma and Orthopedic and Vascular surgery (Supplementary Content, Table SV).
The highest rate of cell salvage use among transfused patients was seen in Cardiac surgery with 9.4% (No.=23,812). The highest rate of massive transfusion (Des.), defined by the Federal Statistical Office as more than 5 RBC in 24 hours, was seen in Thoracic surgery (5.6% - No.=2,617), the lowest in Gynecology (0.85% - No.=93). The highest rate of platelet transfusion was 17% (No.=43,213), seen in Cardiac surgery, the lowest was 1% (No.=11,233) in Trauma and Orthopedic surgery. The highest rate of fresh frozen plasma (FFP) transfusion was 18.1% (No.=45,890), once again in Cardiac surgery. The lowest was 4.9% (No.=531), seen in Gynecology. The most prothrombin complex concentrate (PCC) administration was seen in Neurosurgery with 21.9% (No.=7,112), while the highest rate of fibrinogen administration was seen in Cardiac surgery with 15.6% (No.=39,489). (Table II).
DISCUSSION
In this nationwide analysis of 21,143,317 surgical patients aged ≥65 years, Trauma and Orthopedic surgery had the highest overall RBC transfusion rate and the highest odds ratio for transfusion.
The highest rate of preoperative transfusion was seen in General surgery, while the lowest rate was in Trauma and Orthopedic surgery. These results underline the pivotal role of Trauma and Orthopedic surgery and General surgery in the transfusion management of the elderly. In this context, it seems reasonable to assume that both disciplines might benefit from a targeted optimization and consistent implementation of patient blood management (PBM) strategies to avoid unnecessary blood transfusion and its associated risks.
The German transfusion guidelines were changed in 2020, highlighting patients undergoing Trauma and Orthopedic surgery over 65 with relevant underlying cardiovascular diseases, and setting the transfusion threshold in this particular group at an Hb level of <8 g/dL15. The transfusion threshold in elderly and especially in frail trauma patients is still controversial17,18. In contrast to the high RBC transfusion rate seen in Trauma and Orthopedic surgery, this group of patients also had the lowest rate of platelet transfusion. Since patients who underwent procedures in more than one discipline were excluded, it seems likely that most major trauma cases were not captured in our cohort, and these are the patients which tend to have high rates of massive transfusion and are consequently more likely to receive a platelet transfusion.
The observed increase in transfusion proportions across higher age groups in all surgical disciplines may indicate that patients with rising age show a higher RBC “demand”. While it was impossible to directly assess the degree of frailty within this dataset, the age-related shift emerging in transfusion patterns might well reflect an increasing accumulation of factors associated with diminished physiological reserve in advanced age.
With regard to preoperative RBC transfusion, it was General surgery which showed the highest rates. In this discipline, a high proportion of procedures involve oncologic interventions, particularly for colorectal cancer19, which is often associated with chronic blood loss and preoperative anemia, thereby increasing the likelihood of preoperative transfusion. A study by Papageorge et al. examined preoperative transfusion rates in 60,785 patients undergoing colectomies. They described a lower preoperative transfusion rate of 3.4% (No.=2,073). They performed a univariate analysis and showed that preoperative transfusion is associated with significantly higher rates of postoperative morbidity, and a risk-adjusted, multivariate model showed increased risk of complications with preoperative transfusion20. Studies have shown that perioperative transfusion in patients undergoing Colorectal and Major Abdominal surgery is associated with higher rates of postoperative complications and mortality21,22. Preoperative anemia itself has been identified as an independent predictor of adverse outcomes and increased transfusion rates in Colorectal Cancer surgery23, supporting the conclusion that both -anemia and transfusion-contribute to risk. The high rates of preoperative transfusion observed in General surgery indicate the need for systematic preoperative anemia screening and optimization and should not serve as a justification for prophylactic transfusion. This is consistent with contemporary patient blood management recommendations, which emphasize early detection and treatment of preoperative anemia24.
The RBC transfusion rate in Neurosurgery showed a notable heterogeneity (Supplementary Content, Figure S2). This might reflect the ongoing debate regarding transfusion thresholds in patients with acute brain injury, considering liberal strategies to improve oxygen delivery and reducing brain tissue hypoxia25,26. Our analysis revealed a decrease in the number of patients transfused between 2013 and 2020 which might be associated with the setting-up of the PBM network in Germany27.
The highest platelet transfusion rate was observed in Cardiac surgery, probably due to platelet inhibitors administered preoperatively and the activation due to cardiopulmonary bypass (CPB) resulting in impaired platelet function28 which often leads to post-CPB bleeding and the need for platelet transfusion29,30.
The highest median age of transfused patients (82 years [76;87]) was observed in Trauma and Orthopedic surgery, which is probably due the fact that frailty, osteoporosis, and sarcopenia31 increase the risk of fall and fracture with age32, resulting in the need of major surgery for the treatment of proximal femoral fractures. The distribution within the age groups could be explained by diagnoses typically for the respective age group. Lung cancer is diagnosed in industrialized countries at a median age of 71 years and is the leading cause of death among all malignancies after the age of 80 years33. This might explain the high proportion in patients between 70 and 74 years in Thoracic surgery. Increasing frailty is associated with a higher risk of falls, which may explain the rising proportion of trauma and orthopedic patients in the older age groups.
The Elixhauser-score was higher in transfused patients which indicates that these patients are multimorbid, especially in Thoracic and Neurosurgery. As Blum et al. showed, increasing age and comorbidities are independent risk factors for RBC-transfusion, which explains the higher score among transfused patients1. Among non-transfused patients the highest median Elixhauser-Score was observed in Cardiac surgery, which might be explained by the high share of cardio-vascular risk factors which are listed in this score. The odds ratio (OR) for RBC transfusion exceeded 1.0 in General, Cardiac, Thoracic, Vascular, Trauma, and Orthopedic surgery, indicating an increased likelihood of transfusion in these disciplines. In General surgery, a substantial proportion of procedures involve oncologic interventions; Cardiac, Thoracic, and Vascular surgery involve multimorbid patients (indicated by the high Elixhauser-Score) with underlying cardiovascular diseases which might result in heightened susceptibility to anemia-induced tissue hypoxia. The PBM recommendations from the 2018 Frankfurt Consensus Conference address this by using a different RBC-transfusion threshold (<8 g/dL) in patients with cardiovascular diseases24. In Thoracic surgery specifically, impaired pulmonary function and reduced oxygenation further may contribute to a lower tolerance of anemia. Although the recommendations acknowledge that transfusion thresholds should be adapted to the patient’s physiological reserve and oxygen delivery capacity, chronic pulmonary diseases are not specifically addressed.
The use of intraoperative cell salvage was relatively low, possibly due to concerns regarding reinfusion in Oncologic surgery, which is still the subject of scientific debate and ongoing studies34,35. The Elixhauser Index was higher in transfused patients, particularly in Thoracic and Neurosurgery. This indicates that transfusion decisions in elderly patients may not be prompted solely by blood loss during the procedure but also by patient complexity, comorbidities, and decreased physiological reserves.
The use of legally mandated national reimbursement data in our study ensures a high degree of completeness and standardization: our data set covers a period of ten years and more than 21 million older patients, creating a uniquely broad basis for analysis. The mandatory reporting of ICD-10 and OPS codes, which directly affects hospital reimbursement, should encourage accurate clinical documentation and minimize data gaps. In addition, we excluded cases involving overlapping surgical disciplines in order to improve internal validity and avoid bias due to procedural heterogeneity. While reimbursement data correlate with medical cases in hospitals36, the potential for over- or under-representation of conditions or events for reimbursement purposes cannot be ruled out. Nevertheless, the incentive for accurate documentation is high, given that hospital reimbursement claims are subject to external audits by the medical service of the statutory health insurance funds. Furthermore, the large sample size of over 21 million patients mitigates the potential impact of misclassification or coding errors.
Only surgical disciplines relevant to bleeding and RBC transfusion were analyzed. These disciplines were: General surgery, Gynecology, Cardiac surgery, Thoracic surgery, Oral and Maxillofacial surgery, Neurosurgery, Trauma and Orthopedic surgery, Urology and Vascular surgery. This led to high numbers of patients being excluded, especially among the non-transfused: 73.4% of elderly, surgical non-transfused patients were excluded, while 43% of elderly, surgical transfused patients were excluded. Although this high number is striking, the inclusion of minor surgical procedures, such as in ophthalmology, would certainly have skewed the analysis. Nevertheless, even among the surgical disciplines analyzed, some procedures can be considered minor. To avoid confounding effects from multiple concurrent procedures, patients treated by more than one surgical discipline during the same hospitalization were excluded.
Due to the anonymization progress we were able to analyze age groups divided by 5 years, while studies with similar questions often use the mean age and no breakdown due to much smaller patient groups37. A comparable study by Tinegate et al. which analyzed blood product consumption in the National Health Service (NHS) over one year, used the same surgical disciplines and included cases from internal medicine. They showed similar results in the increase in transfusions with increasing age. The age groups used are identical to ours. It is not possible to compare transfusion rates, as only the proportion of RBC consumption in relation to the total RBC consumption for the year is given38. A secondary analysis of the European Transfusion Practice and Outcome Study (ETPOS)39 examined perioperative transfusion practices. As in our study, the cut-off age was 65. That study analyzed 3,149 patients and provided information regarding the distribution of surgical disciplines, with “Cancer-related surgery” listed as a separate discipline. The most common surgical disciplines in the ETPOS analysis were Orthopedic surgery (34%), General surgery (39%), and Cancer-related surgery (37%). In our study, General surgery accounted for 30% and Trauma and Orthopedic surgery for 42%. The smaller study did not differentiate transfusion rates according to surgical discipline. This underlines the importance of detailed insights and larger cohorts as provided by our study39. The definition of massive transfusion (Des.) as >5 RBC units per patient is the one predefined by the Federal Statistical Office, which reflects an administrative definition and does not correspond to established clinical definitions of massive transfusion. As a result, it is dif ficult to make any direct comparison with studies applying clinical massive transfusion criteria40.
There are some significant limitations to this study which must be acknowledged. The retrospective design and use of administrative data make it impossible to assess the appropriateness of transfusion. Key variables such as hemoglobin levels, clinical bleeding, or intraoperative blood loss, are not coded, which limits any conclusions as regards decision-making or guideline adherence. As a registry study, only associations can be shown. Given the large cohort (No.=21,143,317), all the correlations are statistically significant, but some might be lacking in clinical importance.
The Elixhauser Comorbidity Index, based on binary diagnosis codes, does not capture disease severity, frailty, or functional status. Thus, while multimorbidity is described, physiological reserves and geriatric-specific risks remain unaccounted for. Large-scale data reduce random error but may highlight trivial differences, requiring interpretation based on effect size and plausibility. Finally, logistic regression models yield ORs adjusted for surgical discipline but not for individual risk factors (e.g., anemia, anticoagulant use, procedural complexity). Hence, ORs reflect discipline-level trends rather than individual risk predictions.
CONCLUSIONS
In this nationwide cohort, Trauma and Orthopedic surgery showed the highest overall transfusion rate among elderly patients, while General surgery had the highest proportion of preoperative transfusions. These findings underline the importance of prioritizing evidence-based correction of preoperative anemia rather than relying on prophylactic transfusion strategies. Across all disciplines, the proportion of transfused patients increased with advancing age, indicating an age-dependent rise in transfusion risk, suggesting that advanced age results in reduced physiological capacity. In conclusion, our findings contribute to the ongoing discussion regarding transfusion triggers in older and potentially frail patients, and highlight the need for tailored transfusion strategies in a geriatric patient cohort with a special focus on frailty. In particular, in order to reduce transfusion rates in the elderly, General surgery and Trauma and Orthopedic surgery could benefit from consistent establishment of PBM and perioperative care bundles.
Supplementary Information
Acknowledgements
We would like to thank the Federal Statistical Office for its support and provision of the data.
Footnotes
Ethics approval and consent to participate: Due to institutional anonymization, no conclusions can be drawn about individual patients. According to §21 KHEntgG, the reimbursement data are available for scientific use. The requirement for Ethics Committee approval and informed consent for this study was waived by the Ethics Committee of the University Hospital Frankfurt (Chairman: Prof. Dr. Harder, Ref: 2022-766). All data processing was performed in accordance with the principles of the Declaration of Helsinki.
Consent for publication: As these are anonymized register data, no patient consent could be collected.
Funding and sources: This study was supported by internal institutional research funds from the Department of Anaesthesiology, Intensive Care and Pain Therapy, University Hospital Frankfurt, Goethe University, Frankfurt Germany.
Authorship contributions: LVB and JAK planned the study; TJ conceptualized the data query; JAK performed the statistical analysis and proofreading; LVB and JAK wrote the manuscript with input from FW, CW, SC, KZ and VN. All Authors contributed to the final version.
Disclosure of conflicts of interest: KZ serves as Section Editor in BT but had no role in the peerreview process or editorial decision and has received honoraria for participation in advisory board meetings for Haemonetics and Vifor and received speaker fees from CSL Behring, Masimo, Pharmacosmos, Boston Scientific, Salus, iSEP, Edwards and GE Healthcare. He is the Principal Investigator of the EU-Horizon 2020 project ENVISION (intelligent plug-and-play digital tool for real-time surveillance of COVID-19 patients and smart decision-making in Intensive Care Units) and Horizon Europe 2021 project COVend (Biomarker and AI-supported FX06 therapy to prevent progression from mild and moderate to severe stages of COVID-19). KZ leads as CEO the Christoph Lohfert Foundation as well as the Health, Patient Safety & PBM Foundation. JAK and BF are deputy principal investigators of ENVISION and COVend.
Availability of data and materials
The data on which the results of this study are based are available from the Federal Statistical Office with the restrictions applied. The dataset was used under license for the current study and is therefore not generally accessible. However, the data are available from the corresponding author on reasonable requests and with permission from the Federal Statistical Office.
REFERENCES
- 1.Blum LV, Schmitt E, Choorapoikayil S, Baumhove O, Bayer A, Friederich P, et al. Association of anaemia, co-morbidities and red blood cell transfusion according to age groups: multicentre sub-analysis of the German Patient Blood Management Network Registry. BJS Open. 2022;6(6):zrac128. doi: 10.1093/bjsopen/zrac128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Salive ME, Cornoni-Huntley J, Guralnik JM, Phillips CL, Wallace RB, Ostfeld AM, et al. Anemia and hemoglobin levels in older persons: relationship with age, gender, and health status. J Am Geriatr Soc. 1992;40(5):489–496. doi: 10.1111/j.1532-5415.1992.tb02017.x. [DOI] [PubMed] [Google Scholar]
- 3.Beghé C, Wilson A, Ershler WB. Prevalence and outcomes of anemia in geriatrics: a systematic review of the literature. Am J Med. 2004;116(Suppl 7A):3S–10S. doi: 10.1016/j.amjmed.2003.12.009. [DOI] [PubMed] [Google Scholar]
- 4.Musallam KM, Tamim HM, Richards T, Spahn DR, Rosendaal FR, Habbal A, et al. Preoperative anaemia and postoperative outcomes in non-cardiac surgery: a retrospective cohort study. Lancet. 2011;378(9800):1396–1407. doi: 10.1016/S0140-6736(11)61381-0. [DOI] [PubMed] [Google Scholar]
- 5.Vermeiren S, Vella-Azzopardi R, Beckwée D, Habbig AK, Scafoglieri A, Jansen B, et al. Frailty and the prediction of negative health outcomes: a meta-analysis. J Am Med Dir Assoc. 2016;17(12):1163e1–1163.e17. doi: 10.1016/j.jamda.2016.09.010. [DOI] [PubMed] [Google Scholar]
- 6.Deutsche Gesellschaft für Geriatrie e.V. (DGG) Langversion 1, 1, 2024. AWMF-Registernummer: 084-003. [Accessed on 10/11/2024]. Available at: https://register.awmf.org/de/leitlinien/detail/084-003. [in German]
- 7.Boureau AS, de Decker L. Blood transfusion in older patients. Transfus Clin Biol. 2019;26(3):160–163. doi: 10.1016/j.tracli.2019.06.190. [DOI] [PubMed] [Google Scholar]
- 8.Lubart E, Segal R, Tryhub N, Sigler E, Leibovitz A. Blood transfusion reactions in elderly patients hospitalized in a multilevel geriatric hospital. J Aging Res. 2014;2014:178298. doi: 10.1155/2014/178298. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Goodnough LT, Schrier SL. Evaluation and management of anemia in the elderly. Am J Hematol. 2014;89(1):88–96. doi: 10.1002/ajh.23598. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Simon GI, Craswell A, Thom O, Fung YL. Outcomes of restrictive versus liberal transfusion strategies in older adults from nine randomised controlled trials: a systematic review and meta-analysis. Lancet Haematol. 2017;4(10):e465–e474. doi: 10.1016/S2352-3026(17)30141-2. [DOI] [PubMed] [Google Scholar]
- 11.Kashani HH, Lodewyks C, Kavosh MS, Jeyaraman MM, Neilson C, Okoli G, et al. The effect of restrictive versus liberal transfusion strategies on longer-term outcomes after cardiac surgery: a systematic review and meta-analysis with trial sequential analysis. J Anaesth. 2020;67(5):577–587. doi: 10.1007/s12630-020-01592-w. English. [DOI] [PubMed] [Google Scholar]
- 12.Meybohm P, Lindau S, Treskatsch S, Francis R, Spies C, Velten M, et al. Liberal transfusion strategy to prevent mortality and anaemia-associated, ischaemic events in elderly non-cardiac surgical patients - the study design of the LIBERAL-Trial. Trials. 2019;20(1):101. doi: 10.1186/s13063-019-3200-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Diagnosis-related hospital statistics (DRG statistics) 2005–2021 2023, at (DESTATIS). SÄdBudL. [in German]
- 14.van Walraven C, Austin PC, Jennings A, Quan H, Forster AJ. A modification of the Elixhauser comorbidity measures into a point system for hospital death using administrative data. Med Care. 2009;47(6):626–633. doi: 10.1097/MLR.0b013e31819432e5. [DOI] [PubMed] [Google Scholar]
- 15.Bundesärztekammer [Accessed on 20/10/2025]. Available at: https://www.bundesaerztekammer.de/fileadmin/user_upload/_old-files/downloads/pdf-Ordner/MuE/Querschnitts-Leitlinien_BAEK_zur_Therapie_mit_Blutkomponenten_und_Plasmaderivaten-Gesamtnovelle_2020.pdf. [in German]
- 16.Bundesärztekammer [Accessed on 20/10/2025]. Available at: https://www.bundesaerztekammer.de/fileadmin/user_upload/_old-files/downloads/QLL_Haemotherapie_2014.pdf. [in German]
- 17.Morris MC, Niziolek GM, Baker JE, Huebner BR, Hanseman D, Makley AT, et al. Death by decade: establishing a transfusion ceiling for futility in massive transfusion. J Surg Res. 2020;252:139–146. doi: 10.1016/j.jss.2020.03.004. [DOI] [PubMed] [Google Scholar]
- 18.L’Huillier JC, Hua S, Logghe HJ, Yu J, Myneni AA, Noyes K, et al. Transfusion futility thresholds and mortality in geriatric trauma: Does frailty matter? Am J Surg. 2024;228:113–121. doi: 10.1016/j.amjsurg.2023.08.020. [DOI] [PubMed] [Google Scholar]
- 19.Wong MCS, Huang J, Lok V, Wang J, Fung F, Ding H, et al. Differences in incidence and mortality trends of colorectal cancer worldwide based on sex, age, and anatomic location. Clin Gastroenterol Hepatol. 2021;19(5):955–966e61. doi: 10.1016/j.cgh.2020.02.026. [DOI] [PubMed] [Google Scholar]
- 20.Papageorge CM, Kennedy GD, Carchman EH. Preoperative blood transfusion is a predictor of worse short-term postoperative outcomes after colectomy. Surgery. 2017;161(4):1067–1075. doi: 10.1016/j.surg.2016.08.042. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Jiang T, Liu K, Chen Z. The impact of perioperative red blood cell transfusion on the prognosis of colorectal cancer. Front Surg. 2022;9:927787. doi: 10.3389/fsurg.2022.927787. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Morris FJD, Fung YL, Craswell A, Chew MS. Outcomes following perioperative red blood cell transfusion in patients undergoing elective major abdominal surgery: a systematic review and meta-analysis. Br J Anaesth. 2023;131(6):1002–1013. doi: 10.1016/j.bja.2023.08.032. [DOI] [PubMed] [Google Scholar]
- 23.Mörner ME, Edgren G, Martling A, Gunnarsson U, Egenvall M. Preoperative anaemia and perioperative red blood cell transfusion as prognostic factors for recurrence and mortality in colorectal cancer-a Swedish cohort study. Int J Colorectal Dis. 2017;32(2):223–232. doi: 10.1007/s00384-016-2678-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Mueller MM, Van Remoortel H, Meybohm P, Aranko K, Aubron C, Burger R, et al. Patient Blood Management: recommendations from the 2018 Frankfurt Consensus Conference. JAMA. 2019;321(10):983–997. doi: 10.1001/jama.2019.0554. [DOI] [PubMed] [Google Scholar]
- 25.Taccone FS, Rynkowski CB, Møller K, Lormans P, Quintana-Díaz M, Caricato A, et al. Restrictive vs Liberal Transfusion strategy in patients with acute brain injury: the TRAIN randomized clinical trial. JAMA. 2024;332(19):1623–1633. doi: 10.1001/jama.2024.20424. Erratum in: JAMA 2025; 333(10)911 doi:10.1001/jama.2025.1719. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Taleb C, Gouvea Bogossian E, Bittencour Rynkowski C, Møller K, Lormans P, Quintana Diaz M, et al. Liberal versus restrictive transfusion strategies in subarachnoid hemorrhage: a secondary analysis of the TRAIN study. Crit Care. 2025;29(1):67. doi: 10.1186/s13054-025-05270-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Meybohm P, Hof L, Choorapoikayil S, Zacharowski K. Patient Blood Management: we still have work to do. Transfus Med Hemother. 2023;50(6):561–563. doi: 10.1159/000534087. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Auci E, Vetrugno L, Riccardi I, Brussa A, Orso D, Baroselli A, et al. Multiple Electrode Aggregometry after Cardiopulmonary Bypass to assess Platelet (Dys)-Function and Transfusion Threshold: a concordance study. J Cardiothorac Vasc Anesth. 2020;34(12):3306–3313. doi: 10.1053/j.jvca.2020.06.060. [DOI] [PubMed] [Google Scholar]
- 29.Biancari F, Kinnunen EM, Kiviniemi T, Tauriainen T, Anttila V, Airaksinen JKE, et al. Meta-analysis of the sources of bleeding after adult cardiac surgery. J Cardiothorac Vasc Anesth. 2018;32(4):1618–1624. doi: 10.1053/j.jvca.2017.12.024. [DOI] [PubMed] [Google Scholar]
- 30.Kjellberg G, Holm M, Lindvall G, Gryfelt G, van der Linden J, Wikman A. Platelet function analysed by ROTEM platelet in cardiac surgery after cardiopulmonary bypass and platelet transfusion. Transfus Med. 2020;30(5):369–376. doi: 10.1111/tme.12678. [DOI] [PubMed] [Google Scholar]
- 31.Chen S, Xu X, Gong H, Chen R, Guan L, Yan X, et al. Global epidemiological features and impact of osteosarcopenia: A comprehensive meta-analysis and systematic review. J Cachexia Sarcopenia Muscle. 2024;15(1):8–20. doi: 10.1002/jcsm.13392. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Xu Q, Ou X, Li J. The risk of falls among the aging population: A systematic review and meta-analysis. Front Public Health. 2022;10:902599. doi: 10.3389/fpubh.2022.902599. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Ehrsam JP, Aigner C. [Surgery of old people-Thoracic surgery]. Wien Klin Mag. 2023;26(3):112–121. doi: 10.1007/s00740-023-00497-w. [in German] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Kumar N, Tan JYH, Chen Z, Ravikumar N, Milavec H, Tan JH. Intraoperative cell-salvaged autologous blood transfusion is safe in metastatic spine tumour surgery: early outcomes of prospective clinical study. Eur Spine J. 2023;32(7):2493–2502. doi: 10.1007/s00586-023-07768-4. [DOI] [PubMed] [Google Scholar]
- 35.Aijtink VJ, Rutten VC, Meijer BEM, de Jong R, Isaac JL, Polak WG, et al. safety of intraoperative blood salvage during liver transplantation in patients with Hepatocellular Carcinoma: a systematic review and meta-analysis. Ann Surg. 2022;276(2):239–245. doi: 10.1097/SLA.0000000000005476. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Tyree PT, Lind BK, Lafferty WE. Challenges of using medical insurance claims data for utilization analysis. Am J Med Qual. 2006;21(4):269–275. doi: 10.1177/1062860606288774. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Guo MM, Ji CY, Gu RR, Nan K, Miao CH, Wu QC. Risk factors and outcomes of intraoperative blood transfusion in elderly patients undergoing gastrointestinal cancer surgery. World J Gastrointest Surg. 2025;17(7):105860. doi: 10.4240/wjgs.v17.i7.105860. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Tinegate H, Pendry K, Murphy M, Babra P, Grant-Casey J, Hopkinson C, et al. Where do all the red blood cells (RBCs) go? Results of a survey of RBC use in England and North Wales in 2014. Transfusion. 2016;56(1):139–145. doi: 10.1111/trf.13342. [DOI] [PubMed] [Google Scholar]
- 39.Grüßer L, Keszei A, Coburn M, Rossaint R, Ziemann S, Kowark A ETPOS Study Group. Intraoperative transfusion practices and perioperative outcome in the European elderly: A secondary analysis of the observational ETPOS study. PLoS One. 2022;17(1):e0262110. doi: 10.1371/journal.pone.0262110. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Lin VS, Sun E, Yau S, Abeyakoon C, Seamer G, Bhopal S, et al. Definitions of massive transfusion in adults with critical bleeding: a systematic review. Crit Care. 2023;27(1):265. doi: 10.1186/s13054-023-04537-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
In Germany, hospitals are legally required to report relevant diagnoses using the International Statistical Classification of Diseases and Related Health Problems (ICD) codes and the International Statistical Classification of Procedures (OPS) codes13 (Supplementary Content, Table SI). According to §21 of the German Hospital Finance Law (KHG), hospitals must submit reimbursement data to the Institute for Hospital Remuneration (InEK) for development of the DRG system. These data are forwarded to the Federal Statistical Office (DESTATIS), anonymized, and made available to researchers under a data usage contract. Reporting of ICD-10 and OPS codes, reinforced by reimbursement incentives and monitored by health insurance medical services, ensures high completeness and accuracy. Data are stored locally by DESTATIS, and no patient or hospital identifiers were accessible to the authors. As the dataset was institutionally anonymized, General Data Protection Regulation (GDPR) does not apply. The Ethics Committee of the University Hospital Frankfurt waived the requirement for ethical approval and informed consent (Chair: Prof. Dr. Harder, Ref: 2022-766). Data collection was structured, representative, and conducted in accordance with the Declaration of Helsinki and STROBE guidelines.
The data on which the results of this study are based are available from the Federal Statistical Office with the restrictions applied. The dataset was used under license for the current study and is therefore not generally accessible. However, the data are available from the corresponding author on reasonable requests and with permission from the Federal Statistical Office.