Incidence of major bleeding events and outcome of patients of 80 and 90 years or older with ongoing anticoagulants: five-year survey in northwest Tuscany

Objective

To search for rates of major bleeding events in patients (pts) with age ≥80 or ≥90 years (y.) with ongoing anticoagulants referred to hospital.

Methods

Patients complaining any bleeding events were submitted to propensity score matching for major bleeding and stratified according to age ≥80 or ≥90 y. and warfarin or direct oral anticoagulants (DOACs).

Setting

A General Hospital, northwest Tuscany, five-year survey, 385,650 visits; catchment area 197,722 inhabitants, of whom 18,373 on warfarin and 14,808 on DOACs. Out of DOACs, dabigatran and rivaroxaban were available in the catchment area since 5 y., apixaban 4 y. and edoxaban 3 y; 5,553 pts received rivaroxaban, 4,602 dabigatran, 3,147 apixaban and 1,506 edoxaban.

Endpoint

Primary endpoint was one-week death, and incidence of major bleeding.

Results

Out of 7,474 pts considered, 2504 (33.5%) pts were older than 80 y., of whom 518 (6.8%) were older than 90 y; they were enrolled in the study. Overall, 253 (10.1%) showed history of stroke/TIA, 578 (22.9%) atrial fibrillation, 277 (11.1%) cancer, 177 (7.0%) congestive heart failure, 33 (1.3%) pulmonary thromboembolism.

Of these 7,474 pts 1,040 (41.5%) showed major bleeding: 621 (24.8%) were gastrointestinal of which 258 (10.3%) of the upper tract and 363 (14.5%) of the lower tract; 794 (31.7%) were brain haemorrhage; the remaining patients showed other bleeding.

Overall, 435 (5.8%) pts needed reversal anticoagulation, 325 (4.4%) red blood cell pack, and 2879 (38.5%) admission. Eventually, 127 pts have been readmitted to the hospital for ischemic stroke and 499 for new bleeding event. CHA2D2VASc-score was 2.5±1.5 and Charlston Comorbidity Index was 3.4±2.3.

Out of 2,504 patients older than 80 y., 367 (14,7%) received anticoagulants (including heparin) of which 134 (5.4%) received warfarin versus 63 (2.5%) DOACs (p<0.001); 24 dabigatran, 19 rivaroxaban, 17 apixaban, and 3 edoxaban. Overall 88 (3.5%) needed reversal anticoagulation, 128 pts (5.1%) red blood cell pack, and 825 (32.9%) pts admission.

One-week mortality rate as follows: anticoagulants 35 (1.4%) versus DOACs 6 (0.2%), p<0.001; dabigatran 0, rivaroxaban 2, apixaban 2, edoxaban 2.

Out of 518 patients older than 90 y., 98 (18.9%) received anticoagulants (including heparin) of whom 44 (8.5%) received warfarin; 11 (2.1%) DOACs (p<0.001); 4 dabigatran, 2 rivaroxaban, 4 apixaban, and 1 edoxaban. Overall 24 (4.6%) needed reversal anticoagulation, 50 (9.7%) red blood cell pack, and 203 (39.2%) admission.

One-week mortality rate as follows: anticoagulants 10 (1.9%) versus DOACs 1 (0.2%), p<0.001; dabigatran 0, rivaroxaban 0, apixaban 1 (0.2%), edoxaban 0.

Conclusion

Patients of 80 y. and even 90 y. or older, with ongoing warfarin, showed higher percentage of major bleeding events and mortality rate versus DOACs. Within DOACs, edoxaban was more likely to show lower rate of major bleeding events, without differences in death rate.

Funding Acknowledgement

Type of funding source: None

P657Anticoagulation strategies based on warfarin or direct oral anticoagulants compared to major hemorrhagic events: the relevance of patients aged 75 years or older

Objective

To evaluate rates of major bleeding (mBleed) associated with anticoagulant treatments in patients (pts) with age ≥75 years (y), referred to hospital in northwest Tuscany.

Methods

We analysed 4-y survey; 302,687 visits; catchment area 197,722 inhabitants, of whom 15,267 on Warfarin (W) and 10,397 on direct oral anticoagulants (DOACs). DOACs were available in the catchment area since 4 y (dabigatran and rivaroxaban), 3 y (apixaban), and 2 y (edoxaban). Overall, 3,373 pts received dabigatran, 4,046 rivaroxaban, 2,141 apixaban, and 839 edoxaban. Pts with hemorrhage were submitted to propensity score matching for mBleed and stratified according to age ≥75 y, and W or DOACs. Primary endpoint was one-month death in pts with age ≥75 y.

Results

Out of 1,919 mBleed enrolled, those of pts aged ≥75 y were 1,127 (59%) versus (vs) 792 (41%) aged <75 y, with 77 (4.0%) one-month death vs 20 (1.0%), respectively; p<0.0001.

Patients on W showed higher rate of mBleed (n=175; 9.1%) compared to DOACs (n=53; 2.8%); p<0.0001. One-month death accounted for 14 (0.7%) vs 1 (0.1%), respectively; p=0.0019. Pts aged ≥75 years were more likely to show female gender and comorbidities including atrial fibrillation; p<0.01. Among DOACs, edoxaban presented the lowest absolute rate of hospital admission for mBleed and dabigatran the highest, although without statistical differences within DOACs (p=0.6454). Interestingly, in the subset of pts aged ≥75 y, within DOACs, we found no statistical difference in one-month mortality, although edoxaban showed one death due to brain mBleed; conversely, we found statistical significance in rate of mBleed. Indeed, edoxaban vs dabigatran showed p=0.0008, edoxaban vs apixaban p=0.0242, edoxaban vs rivaroxaban p=0.0058, apixaban vs rivaroxaban p=0.7093, apixaban vs dabigatran p=0.2279, rivaroxaban vs dabigatran p=0.5087.

Warfarin or DOACs and outcomes Major Bleeding (pts ≥75 y) p value versus pts <75 y p value within group One-month death (pts ≥75 y) p value versus pts <75 y p value within group n=1,127 n=77 Warfarin 175 (9.1%) 0.0001 <0.0001 14 (1.2%) 0.514 0.0019 DOACS 53 (2.8%) 0.0001 1 (0.1%) 0.256 Dabigatran 21 (1.1%) 0.026 0.0088 0 (0%) 0.391 0.8012 Rivaroxaban 16 (0.8%) 0.121 0 (0%) 0.619 Apixaban 13 (0.7%) 0.033 0 (0%) 1 Edoxaban 3 (0.2%) 0.647 1 (0.1%) 0.191 Patients: pts; direct oral anticoagulants: DOACs; p value Yates' correction: p value.

Patients with age >75 years and bleeding

Conclusion

In pts with age ≥75 y, rate of mBleed and short-term mortality were significantly higher than in pts aged <75 y In those pts, DOACs showed significantly lower rate of mBleed, and short-term death. Within DOACs, edoxaban was more likely to show lower rate of mBleed.

P4636Major gastrointestinal haemorrhage of patients with ongoing anticoagulants presented to the emergency department of a community hospital: four-year survey

Objective

To look for outcomes of patients (pts) with major gastrointestinal haemorrhage (mGIH) and ongoing anticoagulants out of four-year survey of community hospital with catchment area 197,722 inhabitants, of whom 15,267 with Warfarin (W) and 10,397 with direct oral anticoagulants (DOACs). DOACs were available for prescription in the catchment area since 4 years (dabigatran and rivaroxaban), 3 years (apixaban), and 2 years (edoxaban).

Methods

Haemorrhage (n=1,919) were submitted to propensity score matching for major bleeding; mGIH were enrolled and stratified according to ongoing W or DOACs. Primary endpoint was one-month death.

Results

Out of 476 mGIH, 73 pts received anticoagulants; 22 DOACs and 51 W; p=0.0006. Of note mGIH on W accounted for 2.7% (51/1,919) per year of pts, and 0.08% (51/15,267) of the catchment area. Conversely, mGIH on DOACs accounted as follows: dabigatran (n=10/476) 0.53%, rivaroxaban (n=6/476) 0.32%, apixaban (n=5/476) 0.35%, and edoxaban (n=1/476) 0.11% per year of pts; p=0.117. Rate of mGIH and DOACs versus (vs) rate of mGIH and W as follows: less than (−) 5 fold (2.7x100/0.53) of dabigatran vs W, p=0.004; −8 fold (2.7x100/0.32) of rivaroxaban vs W, p=0.0002; −7 fold (2.7x100/0.35) of apixaban vs W; p=0.ehz745.10188, and −25 fold (2.7x100/0.11) of edoxaban vs W; p=0.ehz745.101801. However no difference versus the catchment area per year (0.723): 0.07% (10/3,373) dabigatran, 0.04% (6/4,046) rivaroxaban, 0.08% (5/2,141) apixaban, 0.06% (1/839) edoxaban.

Overall, one-month death accounted for 10/476 (2.1%). Of note 236 mGI were from the upper tract and 240 from the lower tract. Among upper tract, 10 pts received DOACs (4 dabigatran, 4 rivaroxaban, 2 apixaban, and 0 edoxaban) and 21 received W. One-month death was 0/10 DOACs versus 1/21 W, p=0.483. Among lower tract, 12 pts received DOACs (6 dabigatran, 2 rivaroxaban, 3 apixaban, and 1 edoxaban) and 30 pts received W; one-month death was 0 for every groop. Anticoagulant reversal treatment was given to 6/22 (27%) pts with DOACs versus 18/51 (35%) with W, p=0.014; transfusion to 6/22 (27%) versus 11/51 (22%), respectively, p=0.306; admission 19/22 (86%) versus 42/51 (82%), respectively, p=0.004. Sensitivity/specificity ratio of variables and biomarkers for aggressive pharmacological approach were obtained by area under ROC curve (AUC) >0.50. PTT value >37 sec (AUC 0.57) showed sensitivity 15%, specificity 90%; INR value >1.4 (0.50), sens 15%, spec 80%. In addition, warfarin (0.49) sens 15%, spec 80%; age ≥75 years (0.48) sen 60%, spec 40%.

Gastrointestinal bleeding: flow-chart.

Conclusion

Out of four-year survey, pts with ongoing DOACs were less likely to have mGIH when compared to W.

Patients with W were more likely to receive reversal tratment; pts with DOACs were more likely to undergo admission. Short-term mortality of pts with W was higher than DOACs. Aggressive pharmacological approach should be driven by PTT, INR, ongoing warfarin, and older age.

Role of acetylcholinesterase inhibitors in pharmacological regulation of amyloid precursor protein processing

The triggering events leading to the selective neurodegeneration observed in Alzheimer brains are not yet completely understood. They thus create a great challenge for the definition of a resolutive treatment for the causes and symptoms of Alzheimer's Disease (AD). Since the current therapeutic option for AD patients is the use of acetylcholinesterase inhibitors (AChEIs), several authors have examined whether these drugs can also affect the expression and metabolism of the amyloid precursor protein (AbetaPP). The rationale behind these studies was based on the fact that the literature suggests that cholinergic activities are also involved in the regulation of AbetaPP metabolism. Therefore, the characterization of these aspects of AD pharmacology may allow cholinergic drugs to be tested for their ability to intervene at different levels of the pathogenetic chain, other than providing a replacement therapy for lost neurotransmitters. This paper reviews the evidence that many of these drugs, although with different qualitative effects, are able to modulate the metabolism and expression of AbetaPP. This effect is often sustained by an indirect cholinergic mechanism and does not affect the mRNA expression of the precursor, although some other authors have demonstrated an effect on post-transcriptional regulation of AbetaPP expression. In addition to the effect on AbetaPP processing, we recently explored the possibility that these molecules affect a gene expression program beyond the classical pharmacological effects, for insights on possibly unexplored pathways of intervention in AD.

Role of acetylcholinesterase inhibitors in the regulation of amyloid β precursor protein (AβPP) metabolism

The events that lead to neurodegeneration in Alzheimer's brains are largely unknown and this fact creates a great challenge for the definition of treatments that may be resolutive for Alzheimer's disease (AD). The current therapeutic option for AD patients is the use of acetylcholinesterase inhibitors (AChEIs), which gives a symptomatic relief to some of the clinical manifestations of the disease. In addition, several authors investigated whether these drugs can also affect one of the major pathogenetic pathway postulated for the disease, that is the expression and metabolism of the amyloid precursor protein (AbetaPP). The literature suggests that cholinergic activities may be significantly involved in the regulation of AbetaPP metabolism, thus the characterisation of these aspects of AD pharmacology may allow testing cholinergic drugs in their ability to intervene at different levels of the pathogenetic chain, other than to provide a replacement therapy for lost neurotransmitters. In this paper we review the evidence that these drugs, albeit with different quantitative and qualitative effects, can modulate the metabolism and expression of AbetaPP, through mechanism that involve either an indirect cholinergic mechanism or effects that do not involve the canonical pharmacological activity of AChE inhibitors. We also provide preliminary evidence that these molecules may affect a gene expression program beyond the classical pharmacological pattern suggesting an insight on possible unexplored pathways of intervention in AD.

Acetylcholinesterase inhibitors: novel activities of old molecules

The therapeutic approach for improving the cognitive function in patients with Alzheimer's disease (AD) is mainly based on the potentiation of central cholinergic activity and is achieved clinically by the use of acetylcholinesterase (AChE) inhibitors such as tacrine, donepezil, rivastigmine, galantamine and other drugs currently in clinical trials. These are, by their pharmacology, only symptomatic drugs yet recently these molecules have shown some potential also in the modulation of amyloid precursor protein (APP) processing. We explore in this review the experimental evidence that suggests a role for AChEIs in APP processing and point to multiple complex mechanisms involving either a cholinergic agonist effect, coupled to multiple signal transduction pathways, or post-transcriptional effects that modulate the expression of cellular APP.

Differential involvement of protein kinase C alpha and epsilon in the regulated secretion of soluble amyloid precursor protein

We investigated the differential role of protein kinase C (PKC) isoforms in the regulated proteolytic release of soluble amyloid precursor protein (sAPPalpha) in SH-SY5Y neuroblastoma cells. We used cells stably transfected with cDNAs encoding either PKCalpha or PKCepsilon in the antisense orientation, producing a reduction of the expression of PKCalpha and PKCepsilon, respectively. Reduced expression of PKCalpha and/or PKCepsilon did not modify the response of the kinase to phorbol ester stimulation, demonstrating translocation of the respective isoforms from the cytosolic fraction to specific intracellular compartments with an interesting differential localization of PKCalpha to the plasma membrane and PKCepsilon to Golgi-like structures. Reduced expression of PKCalpha significantly impaired the secretion of sAPPalpha induced by treatment with phorbol esters. Treatment of PKCalpha-deficient cells with carbachol induced a significant release of sAPPalpha. These results suggest that the involvement of PKCalpha in carbachol-induced sAPPalpha release is negligible. The response to carbachol is instead completely blocked in PKCepsilon-deficient cells suggesting the importance of PKCepsilon in coupling cholinergic receptors with APP metabolism.

Characterization of the effect of ganstigmine (CHF2819) on amyloid precursor protein metabolism in SH-SY5Y neuroblastoma cells

We have investigated the effect of ganstigmine (CHF2819), a novel geneserine derived acetylcholinesterase (AChE) inhibitor, on the expression and metabolism of the amyloid precursor protein (APP) in neuroblastoma cell line SH-SY5Y. The rationale was based on the suggestion that cholinergic activity may also be involved in the regulation of APP metabolism. We studied the acute effect on APP metabolism following the secretion of sAPPalpha in the conditioned medium of cells. Following short term treatment (2h), ganstigmine promoted a slight increase in the release of sAPPalpha, the maximal effect approaching on average 1.5 fold baseline value. The data obtained in the long term experiments demonstrate that continuous inhibition of AchE obtained with 100 nM ganstigmine following an exposure of 24 hours did not influence APP isoforms expression. However, the compound appeared to increase the constitutive release of sAPPalpha, with a mechanism that is derived from an indirect cholinergic stimulation.

Role of protein kinase Calpha in the regulated secretion of the amyloid precursor protein

Protein kinase C (PKC) has a key role in the signal transduction machinery involved in the regulation of amyloid precursor protein (APP) metabolism. Direct and indirect receptor-mediated activation of PKC has been shown to increase the release of soluble APP (sAPPalpha) and reduce the secretion of beta-amyloid peptides. Experimental evidence suggests that specific isoforms of PKC, such as PKCalpha and PKC epsilon, are involved in the regulation of APP metabolism. In this study, we characterized the role of PKCalpha in the regulated secretion of APP using wild-type SH-SY5Y neuroblastoma cells and cells transfected with a plasmid expressing PKCalpha antisense cDNA. Cells expressing antisense PKCalpha secrete less sAPPalpha in response to phorbol esters. In contrast, carbachol increases the secretion of sAPPalpha to similar levels in wild-type cells and in cells transfected with antisense PKCalpha by acting on APP metabolism through an indirect pathway partially involving the activation of PKC. These results suggest that the direct PKC-dependent activation of the APP secretory pathway is compromised by reduced PKCalpha expression and a specific role of this isoform in these mechanisms. On the other hand, indirect pathways that are also partially dependent on the mitogen-activated protein kinase signal transduction mechanism remain unaffected and constitute a redundant, compensatory mechanism within the APP secretory pathway.