Patients taking DOACs are a growing subpopulation of patients presenting with PFF. Choosing a safe time frame for surgery is crucial and should be based on sound data analysis specific to this patient subgroup. The results of this study support early surgical management of patients with a proximal hip fracture, regardless of DOAC status.
Literature to date has reported mixed data on transfusion need in DOAC patients, with no clear relationship to time to surgery7,8. In our overall sample, blood transfusions were needed more often when time to surgery increased. This effect was highlighted in DOAC patients, where 38% needed a transfusion when operated after 24 hours compared to 18% of those operated on earlier. This suggests a similar perioperative blood loss between DOAC patients and controls. Moreover, it provides insight that other factors may be influencing transfusion rate more than DOAC status, which is inconsistent with blanket statements that time to surgery should be increased for DOAC patients. This is an effect that has been previously reported on, most recently by Ashley et al in 132 patients taking DOACs9. Similarly, perioperative hemoglobin change was relatively stable between 2,4 and 2,7 g/dL throughout all performed analyses. Similar hemoglobin change, irrespective of DOAC status or time to surgery, also undermines the benefit of delaying surgery for DOAC patients. This effect is in line with the current literature10–13.
Many studies that examined the relationship between perioperative blood loss and time to surgery have stressed that early surgery leads to less transfusion need in DOAC patients. Increased blood loss is the stated possible negative effect of later surgery. From the orthopedic trauma perspective this is self-evident, as perioperative blood loos in hip fractures can be up to six times higher than visible intraoperative blood loss14. Therefore, more factors must be taken into consideration instead of only anticoagulation status. Moreover, past studies have shown that BMI, time to surgery and fracture type have more impact on blood loss in proximal femur fracture than anticoagulation status15. Our results could not establish a significant relationship between surgical delay and hemoglobin change, in line with previous results by Mullins et al.16.
Length of postoperative stay was not significantly longer in DOAC patients compared to controls. When however, examining its relationship to time to surgery in DOAC-patients, a clear benefit of six days was seen in patients who received operative care within 24 hours (p = 0.0167), the effect being stronger than the analysis of the whole population. This is an indication that the well-known benefits of early operative management in geriatric hip fractures might be amplified in patients taking DOACs. King et al reported similar effects of time to surgery on postoperative length of stay. The study of 28 DOAC patients shows supportive data of early (< 48h) surgery as a delay longer than 48 hours resulted in a mean lengthening of seven days17. Other studies have unfortunately reported on total length of stay, rather than length of postoperative stay and therefore cannot be easily compared.
A positive correlation between time to surgery and postoperative length of stay was established in the entire patient population, though its effect could not be repeated on the isolated DOAC group. This is most likely due to an insufficient sample size, as the effect was closer to statistical significance with a steeper slope than the regression analysis in the control subgroup, despite larger size. (p = 0.10 for n = 59 vs p = 0.23 for n = 255).
Our DOAC patients had a similar rate of combined medical and surgical complications compared to controls. Treatment before 24 hours was beneficial for all patients reducing the complication rate by 14% (p = 0.0167). A similar effect was seen in the DOAC subgroup, where earlier surgery resulted in a 11% decrease in complication rate (p > 0.05). Data on medical complications after early operative management in DOAC patients has not previously been reported on. Moreover, the number of complications was recorded rather than the number of patients who suffered a complication. This gains better insight into the true resources that are needed to treat these patients.
The two-year revision rate was similar between DOAC patients and controls, although DOACS patients had significantly more revisions for hematomas. When looking at DOAC patients separately, a longer time to surgery does not decrease revision rate and an overall decrease in revisions is seen when operated on withing 24 hours. 75% of hematomas needing revision were initially operated on after 24 hours and thus were not prevented by waiting longer than 24 hours. This suggests that waiting is not the treatment of choice to avoid revision surgery. Other authors have previously reported on 30-day revision rate. Franklin et al has no revisions in the DOAC subgroup after 30 days, while Mullins et al reports a 5% revision10,16. The difference in revision rate in our series can be partially explained by the inclusion of implant removals and the longer follow up period of two years. When these are excluded, the overall two-year revision rate is 7,6%. When only considering hematoma and infection, the revision rate is 5,5% respectively.
The mean waiting time to surgery was significantly longer in DOAC patients, where operations started 7 hours and 17 minutes after their peers. This is an effect, that has been shown by multiple researchers10,12,18,19. This is most probably due to the careful stance in current guidelines on perioperative management of DOACs4,20.
Current data on trauma patients with coagulation disorders, either acquired or congenital, generally shows an elevated mortality risk in patients with or without head trauma. Even as this data is valid it should be interpreted with caution when managing risk in elderly PFF. Often, data represents either polytraumatized patients or does not differentiate between coumarins and DOAC patients. Due to the novel nature of the medication, data collected before 2008 cannot be used to estimate risk in DOACs patients, as this is before FDA and EMA approval of these medications. It is the authors opinion that the current stance towards waiting for passive reversal for hip fracture surgery in low energy, monotraumatized patients taking DOACs is not supported by prospective data. Therefore, the bleeding risk in this patient group might be overestimated due to the lack of specific data to prove the contrary. Moreover, a recent study by Bläsius et al. based on the German Trauma Registry®, showed a higher mortality in patients taking coumarins (677 pairs; 23,9% vs 19,5%; p = 0.026), but these results could not be repeated for patients taking DOACs (437 pairs, 19,0 vs 19,7%; p = 0.84)23.
Morbidity and mortality data in anticoagulated patient in the event of polytrauma, head or visceral trauma does not reflect the perioperative risk of hip fracture surgery. Clinicians must beware of overgeneralization of trauma patients as it might affect patients with monotrauma negatively. A growing body of research is showing that early hip surgery is safe in patients taking DOACs and should not be deliberately delayed10,16,18. Furthermore, displaced fractures can cause extensive bleeding without timely and sufficient operative treatment. As the early surgical treatment of PFF benefits patients so clearly, the adverse effect of operating early with DOACs should be at least equally negative to support a watchful waiting practice as it is known in elective surgery.
The strength of the study lies in the fact that it is the first that collects data on medical complications in this patient group. In addition, the two-year revision rate has not been reported on in previous studies in the patient population. A source of bias could involve the fact that generally healthier patients would be approved for surgery faster, regardless of DOAC intake. This could positively skew the results of the earlier treatment group because their general health was better before trauma.
In absence of orthopedic trauma guidelines for DOAC-patients with a hip fracture, waiting for the coagulative effect to wear off as it is preferred in elective surgery could welcome the well-researched increased risk of mortality in this frail population, even though bleeding complications are of the theoretical kind. These effects have been thoroughly investigated and proven to such an extent, that many countries have active legislation to discourage unnecessary delays in treatment with “best tariff” reimbursement, fines and legal procedures.