Postpartum hemorrhage (PPH) remains a major cause of maternal morbidity and mortality. This single‑center retrospective observational study evaluated how local structural changes — first the establishment of a hemostaseology working group and later the clinical implementation of the interdisciplinary D‑A‑CH algorithm — influenced PPH management at the University Hospital Ulm between March 2003 and July 2021.
The primary aim was to assess changes in targeted hemostatic therapy (use and dose of fibrinogen concentrate and tranexamic acid) and transfusion of allogeneic blood products. Secondary aims included use of additional interventions and diagnostics and effects on intensive care unit (ICU) and total hospital length of stay.
Patient records for all women treated for PPH during the specified period were reviewed (n = 341). Inclusion criteria included estimated blood loss thresholds of at least 500 ml after vaginal delivery or 1000 ml after cesarean section. Patients were divided into three chronological groups corresponding to institutional changes in PPH care:
The study compared hemostatic therapies, transfusion needs, use of adjunctive measures, and clinical outcomes across these groups.
The three groups did not differ significantly in key baseline variables reported by the authors: patient age, body mass index (BMI), parity, mode of delivery, or estimated blood loss. This similarity supports interpretation of temporal changes as related to management changes rather than obvious differences in case mix.
Following institutional changes, the study observed marked differences in targeted coagulation therapy. Implementation of the hemostaseology working group alone (Group II) was associated with increased administration of fibrinogen concentrate and tranexamic acid compared with Group I. After the D‑A‑CH algorithm was implemented (Group III) these increases were larger and statistically significant:
These results indicate more proactive, targeted hemostatic therapy after algorithm adoption.
Concurrently with increased targeted hemostatic therapy, use of allogeneic blood products declined after the D‑A‑CH algorithm was implemented. Reported median units of concentrated red blood cells were:
This reduction in red cell transfusion reached statistical significance (p = 0.0036). The authors note a similar significant reduction in fresh frozen plasma use after algorithm implementation.
After the working group and algorithm were introduced, several adjunctive measures and diagnostics were used more frequently. The study reports significant increases in the use of sulprostone and viscoelastic coagulation testing after both interventions. Following algorithm implementation, the Bakri balloon tamponade was also used more often. These shifts reflect broader changes in the institutional PPH treatment approach toward both pharmacologic and mechanical measures.
Clinical course metrics improved after D‑A‑CH implementation:
Duration of mechanical ventilation did not differ between groups. Rates of mortality and acute kidney injury were reported as rare and showed no meaningful change across study periods.
Key between‑group differences reported as statistically significant include the increased administration and dosing of fibrinogen and tranexamic acid (both p < 0.0001), reduced allogeneic blood product use (p = 0.0036), and shorter ICU and hospital stays (p = 0.0195 and p < 0.0001, respectively). The authors also state that reductions in blood product consumption remained after adjustment for relevant confounders, although specific adjustment variables and detailed multivariable results are not reproduced here beyond the abstract.
The authors emphasize that the retrospective, single‑center design prevents definitive causal inference. While the temporal association between D‑A‑CH algorithm implementation and improved targeted hemostatic therapy, reduced transfusion requirements, and shorter stays is clear in this cohort, causation cannot be established from these data alone. The study supports the feasibility and potential benefit of structured, interdisciplinary PPH algorithms and of specialized hemostaseology support, and the authors advocate for further research to confirm these findings.
Overall, the report documents a shift toward earlier, targeted administration of fibrinogen and tranexamic acid, broader use of viscoelastic diagnostics and mechanical measures, and associated reductions in transfusion and length of stay after adoption of the D‑A‑CH algorithm at this level‑1 perinatal center.