SSRI-induced coagulopathy: is it reality?

Abstract

Objectives: To study the effect of escitalopram and fluoxetine on coagulation profile in patients with major depression.

Method: This was a prospective, open-label, single-centre study in 40 patients diagnosed with major depressive disorder. The patients were diagnosed using Diagnostic and Statistical Manual of Mental Disorders, fourth edition criteria. Twenty patients receiving escitalopram 10 mg per day and 20 patients receiving fluoxetine 20 mg per day participated in the study and were followed up for 3 months. Coagulation parameters – bleeding time, clotting time, platelet count, prothrombin time and partial thromboplastin kaolin time – were evaluated at baseline and after 3 months.

Results: At the end of 3 months, a significant increase in bleeding time was seen in patients receiving fluoxetine, but within the normal range. No rise was seen in the group given escitalopram.

Conclusion: In patients with depression, fluoxetine increases bleeding time whereas escitalopram has no effect on coagulation profile. However, both the drugs can be used safely for long-term treatment.

Keywords

coagulopathy escitalopram fluoxetine SSRI

Introduction

Selective serotonin reuptake inhibitors (SSRIs) are widely prescribed for the treatment of depression, obsessive compulsive disorders, bulimia, generalized anxiety and phobic disorders. The majority of all antidepressants prescribed worldwide are from the SSRI family. Commonly prescribed SSRIs include fluoxetine, paroxetine, sertraline, citalopram, escitalopram and fluvoxamine. Unlike tricyclic antidepressants, SSRIs do not have anticholinergic side effects and are safe in overdose [Rang et al. 2007]. Common adverse events are gastrointestinal side effects, sexual dysfunction, headaches, anxiety, insomnia and sedation. There are reports of increased incidence of epistaxis and ecchymosis with SSRIs, which is probably due to impairment of platelet function. Gastric blood loss caused by nonsteroidal anti-inflammatory drugs (NSAIDs) may be increased by SSRIs. Bleeding events are rare but there may be potentially severe haematological complications following treatment with SSRIs. Fluoxetine, a commonly used SSRI, has been reported to cause ecchymosis, bleeding and other haematological complications. In a single case report, a 23-year-old woman treated with fluoxetine for 10 weeks reported ecchymosis and bleeding without any trauma. Her coagulation parameters were found to be normal and the ecchymosis disappeared after the medication was discontinued for 4 weeks [Mirsal et al. 2002].

In a case control study by Douglas and colleagues, no association was found between SSRI use and hemorrhagic stroke [Douglas et al. 2011]. Drugs with the highest level of serotonin reuptake inhibition, such as fluoxetine, paroxetine and sertraline, are more frequently associated with abnormal bleeding [Halperin and Reber, 2010].The most frequent haemostatic abnormalities are decreased platelet aggregation and prolongation of bleeding time.

Epidemiological evidence on the association between SSRI use alone and in combination with NSAIDs indicates that SSRI use might play a causal role in upper gastrointestinal bleeding and that it might act synergistically with NSAIDs or low-dose aspirin. Antidepressants having no effect on the serotonin receptor had no significant effect on the risk of upper gastrointestinal bleeding [Dalton et al. 2003, 2006].

A prescription-event monitoring database in England analysed combined haemorrhagic event rates calculated for the first 6 months of treatment with four SSRIs. The database showed no significant difference between the rates of abnormal bleeding in the first month after treatment when compared with 2–6 months [Layton et al. 2001].

Although studies with equivocal results are limited, there is a sense of caution in the clinical practice of psychiatry, particularly in patients with depression with hematological disorders and patients on SSRIs undergoing major surgical procedures. Therefore, the present study was undertaken to investigate the effects of SSRIs on the coagulation profile of patients with major depression.

Method

A prospective, open-label study was conducted at the psychiatry outpatient department of a tertiary care hospital. The study started after seeking permission from the institutional ethics committee. Written informed consent was taken from all the patients prior to study procedures. Male and female patients aged between 19 and 65 years who were suffering from major depressive disorder according to Diagnostic and Statistical Manual of Mental Disorders, fourth edition [American Psychiatric Association, 2000] were included. Patients not capable of giving consent, patients with comorbid psychiatric illnesses, patients with comorbid medical or surgical illnesses, patients receiving any medication apart from antidepressants and pregnant women were excluded.

Twenty patients were receiving escitalopram (10 mg/day) and 20 patients were receiving fluoxetine (20 mg/day) without any further increase in the dose of drugs. Baseline blood samples were collected before starting the treatment to measure the coagulation profile. These patients were receiving cognitive therapy along with pharmacotherapy. Patients were followed up every week until 3 months for adverse drug reactions. Patients were reassessed after 3 months of treatment. Blood samples were again collected and checked for any change in coagulation profile. Laboratory testing was done in the institutional laboratory. Coagulation profile comprised bleeding time (measured by Ivy’s method) [Dacie and Lewis, 1975a] and clotting time (measured by Lee and White’s method) [Dacie and Lewis, 1975b]. Platelet count was determined using an automated cell counter, and prothrombin time and partial thromboplastin kaolin time (PTTK) were measured by a coagulometer.

A bleeding time of 2–7 min, a clotting time of 4–9 min, a platelet count of 1.5–4 lacs/mm³, a prothrombin time of 11–16 s (as per the control) and a PTTK of 30–40 s were considered normal. The data obtained were analysed using descriptive statistics and paired Student’s t-test to compare the results from baseline.

Statistical analysis

This was a preliminary, exploratory study to assess the effects of escitalopram and fluoxetine on coagulation profile in patients with major depression.

Patient data were analysed on an intent-to-treat basis in Excel. Nominal data (e.g. sex) were expressed as number and percentage, and continuous data (e.g. age and coagulation parameters) as mean and standard deviation (SD). Paired Student’s t-test was used for a within-group (pre versus post) comparison. A p value of 0.05 or less was considered statistically significant.

Results

In both study groups, eight (40%) patients were men and 12 (60%) were women. The average age was 32 ± 10.89 years in the escitalopram group and 31.95 ± 9.45 years in the fluoxetine group.

The coagulation profile for patients receiving escitalopram is given in Table 1. In the escitalopram group, no significant differences in coagulation parameters were observed when compared with baseline.

Table 1.

Coagulation profile for patients receiving escitalopram.

	Baseline					After 3 months
Patient no.	BT	CT	PC	PT (t/c)	PTTK	BT	CT	PC	PT (t/c)	PTTK
1	1.30	4.45	2.57	13.6 (13)	33.0	1.40	4.48	3.68	13.4 (14)	30.0
2	1.40	5.28	3.00	14.2 (14)	31.4	1.30	5.30	2.92	14.2 (14)	30.0
3	1.30	5.15	2.86	12.6 (12.2)	30.0	1.40	5.20	2.82	13.2 (13.1)	32.0
4	2.00	5.30	1.69	15.2 (15)	33.1	2.12	5.20	3.28	15 (14.8)	32.0
5	1.45	5.30	3.20	14 (14.5)	31.0	1.60	5.15	3.10	14.5 (13.9)	31.0
6	2.82	4.80	3.45	14 (14)	31.0	2.90	4.90	3.40	15 (15)	31.0
7	1.86	4.20	2.32	13.5 (13.6)	32.20	1.96	4.42	2.30	12.3 (12)	31.0
8	3.26	5.12	3.28	15.5 (15)	31.00	3.30	5.32	3.12	13.5 (14)	30.00
9	2.92	4.90	1.90	12.4 (12)	31.40	3.00	4.98	1.86	16 (15)	31.50
10	3.86	4.80	3.80	15.4 (15.8)	32.00	3.90	4.68	3.72	13.2 (13)	32.00
11	2.90	4.86	3.20	11.6 (11)	30.00	2.92	4.90	3.10	12.3 (12.2)	31.00
12	3.20	5.00	2.86	15 (14.8)	31.00	3.32	5.00	2.70	14.6 (14.1)	31.40
13	2.12	4.20	2.96	16 (15)	30.00	2.22	4.32	1.78	15.3 (14.9)	30.00
14	2.20	5.86	2.10	14 (3)	32.00	2.26	5.90	2.28	12.5 (12.1)	32.20
15	1.86	4.86	2.32	11.9 (11)	31.00	1.96	4.80	2.30	13.2 (13)	31.10
16	3.30	4.65	3.44	14.2 (14)	31.40	3.10	4.60	3.26	14.6 (14.2)	31.00
17	2.86	5.12	2.00	13.8 (13.2)	32.00	2.60	5.00	1.90	13.5 (13)	32.00
18	3.12	5.50	2.26	12.5 (12.1)	32.60	3.26	5.56	2.24	13.2 (12.3)	33.00
19	3	4.18	3.20	14.6 (14)	32.00	2.96	4.65	3.10	14.2 (13.8)	33.00
20	2.96	5.22	3.48	15 (14.6)	31.00	3.00	5.45	3.44	15.2 (15)	32.00
Mean	2.48	4.94	2.79	13.94	31.46	2.52	4.99	2.82	13.95	31.36
SD	0.78	0.45	0.61	1.14	0.91	0.75	0.41	0.61	1.06	0.93
Parameters	Baseline					After 3 months				p value
BT (min)	2.48 ± 0.78					2.52 ± 0.75				0.13
CT(min)	4.94 ± 2.79					4.99 ± 0.41				0.12
PC (lacs/mm³)	2.79 ± 0.61					2.82 ± 0.61				0.51
PT (s)	13.94 ± 1.14					13.95 ± 1.06				0.73
PTTK (s)	31.46 ± 0.91					31.36 ± 0.93				0.70

Data were analysed by paired Student’s t-test. Each value is the mean ± SD.

BT, bleeding time; CT, clotting time; PC, platelet count; PT (t/c), prothrombin time (test/control); PTTK, partial thromboplastin time with kaolin; SD, standard deviation.

The coagulation profile for patients receiving fluoxetine is given in Table 2. In this group, there was a significant increase in the bleeding time after 3 months of treatment compared with baseline, but this was not beyond the normal range as seen in the table. For the other parameters – clotting time, platelet count, prothrombin time and PTTK – no significant differences were observed.

Table 2.

Coagulation profile for patients receiving fluoxetine.

	Baseline					After 3 months
Patient no.	BT	CT	PC	PT (t/c)	PTTK	BT	CT	PC	PT (t/c)	PTTK
1	2.30	4.45	3.28	12.4 (13)	32.00	2.40	4.60	3.12	12.6 (12)	32.00
2	1.86	5.80	2.80	13.4 (13)	31.00	3.12	5.90	2.62	13.2 (13)	32.00
3	3.86	6.20	3.21	14 (13.2)	32.00	3.90	4.20	3.12	13.9 (14)	31.00
4	2.60	4.96	1.90	13.3 (12.5)	31.40	2.60	5.10	1.86	13.8 (13.2)	31.20
5	3.12	5.00	2.94	13.2 (13)	32.20	3.52	5.20	1.86	14.2 (14.1)	31.20
6	2.86	4.89	2.86	15.7 (14.8)	31.00	2.90	4.90	2.76	13.2 (14.1)	32.00
7	2.30	4.90	1.82	15 (14)	32.20	2.34	5.14	1.80	15.3 (15)	31.00
8	3.12	5.20	3.30	12.8 (12)	31.40	3.32	4.12	3.18	12.6 (12.8)	31.20
9	1.98	5.10	2.80	14.3 (14)	30.00	3.16	5.20	2.60	14.4 (14)	32.10
10	2.36	4.98	2.96	13.5 (14)	33.10	2.46	5.00	3.00	14 (13.5)	33.40
11	2.30	4.20	2.80	14.1 (13.8)	30.00	3.20	4.62	2.82	14.6 (14)	31.00
12	2.40	4.32	2.60	11.9 (12)	30.00	3.20	4.86	2.56	12 (11.5)	31.20
13	2.20	4.30	3.90	15 (14)	31.00	2.30	4.32	3.86	15.3 (15)	31.20
14	2.40	4.75	2.55	13.8 (13.2)	31.00	2.52	5.00	2.60	14.3 (14.1)	31.30
15	2.96	5.30	3.12	14.6 (14.2)	32.00	3.00	4.20	3.00	14.9 (14)	32.40
16	3.12	4.96	2.20	15.7 (15)	32.10	3.20	4.86	2.16	15.5 (15)	32.40
17	2.76	5.26	3.92	14.3 (14.1)	31.00	2.86	5.30	3.85	15 (14.8)	32.00
18	3.00	5.50	1.92	14.2 (13.2)	32.00	3.42	5.60	2.00	13 (12.2)	32.20
19	2.12	4.68	2.25	12.6 (12)	31.00	2.20	4.36	3.32	13.2 (13)	32.00
20	3.12	4.86	1.94	11.5 (11.4)	30.00	2.12	4.90	1.90	12 (11.2)	30.00
Mean	2.64	4.98	2.75	15.83	31.32	2.89	4.87	2.70	13.85	31.64
SD	0.50	0.49	0.62	0.59	0.88	0.50	0.48	0.63	1.09	0.74
Parameters	Baseline					After 3 months				p value
BT (min)	2.59 ± 0.50					2.84 ± 0.49				0.03*
CT (min)	4.98 ± 0.49					4.87 ± 0.48				0.42
PC (lacs/mm³)	2.75 ± 0.62					2.70 ± 0.63				0.50
PT (s)	15.83 ± 0.65					15.88 ± 0.65				0.62
PTTK (s)	31.32 ± 0.88					31.64 ± 0.74				0.09

Data were analysed by paired Student’s t-test. Each value is the mean ± SD.

Statistically significant.

BT, bleeding time; CT, clotting time; PC, platelet count; PT, prothrombin time; PTTK, partial thromboplastin time with kaolin.

Discussion

Life-time risk of major depression is 5–10% and is twice as common in women compared with men [Baldessani et al. 2006]. SSRIs are one of the most widely used drugs for the treatment of depression. They are well tolerated and have fewer side effects than older tricyclic antidepressants and are thus preferred [Rang et al. 2007].

A recent increased incidence of epistaxis and ecchymosis with SSRI use has been reported, probably because of impairment of platelet function. Serotonin is one of the mediators released during platelet release reaction, causing platelet aggregation. In one study, five children aged between 8 and 15 years developed bruising or epistaxis 1 week to 3 months after starting SSRI treatment. It is possible that the impact of SSRIs on platelet function are causing these effects or a separate coagulopathy exists in these patients [Lake et al., 2000]. Gastric blood loss due to NSAIDs can be increased by SSRIs [Dalton et al. 2003, 2006; Weinreib et al. 2005; Wessinger et al. 2006]. Although bleeding events are rare, there can be potentially severe haematological complications following treatment with SSRIs in patients with major depression [Mirsal et al. 2002].

A literature search has revealed that SSRI use alone or in combination with other synergistic drugs can cause increased bleeding episodes, including upper gastrointestinal bleeding [Dalton et al. 2003, 2006; Weinreib et al. 2005; Wessinger et al. 2006; Schalekamp et al. 2008; Kumar et al. 2009; Andrade et al. 2010; Strubel et al. 2010]. In our study, fluoxetine caused an increase in bleeding time after 3 months of treatment compared with the baseline values, but this increase was not beyond the normal range of bleeding time. This is in accordance with the study by Halperin and Reber [Halperin and Reber, 2010]. There was no significant difference in other coagulation parameters with fluoxetine after 3 months of treatment.

In the escitalopram group, no significant difference was seen in the coagulation profile after 3 months of treatment. The reason could be that fluoxetine is a more powerful inhibitor of serotonin reuptake compared with escitalopram [Halperin and Reber, 2010]. Adverse effects like decreased appetite, bowel disturbances and insomnia were seen in both groups.

Fluoxetine was found to significantly affect the bleeding time but the increase was not beyond the normal range. The risk of SSRI-associated gastrointestinal bleeding is increased with the concurrent use of NSAIDs, anticoagulants and antiplatelet agents and is decreased by concurrent proton pump inhibitors. The risk of bleeding is increased in patients with cirrhosis of the liver or liver failure. There is little literature on the use of SSRIs and menstrual or postpartum blood loss. Maternal SSRI intake is not associated with an increase in bleeding time in neonates [Maayan-Metzger et al. 2006]. In this study, none of the patients received any drugs apart from SSRIs, hence it is difficult to comment on whether SSRIs increase the risk of bleeding when used in combination with NSAIDs. None of the women (60% of the sample) reported any change in usual menstrual flow with escitalopram or fluoxetine. Pregnancy was ruled out when women were included in the study.

In one case report, citalopram caused severe thrombocytopenia leading to haemorrhage after 4 weeks of treatment and its withdrawal led to patient recovery [Andersohn et al. 2009]. SSRIs appear to be protective against ischemic heart disease events. The data are too limited to examine the influence on ischemic and hemorrhagic stroke [Ramasubbu, 2004; Andrade et al. 2010]. More studies are required in this field with more specific tests of platelet function.

Considering the conflicting reports from numerous studies and the findings from this study, the benefit of using SSRIs in patients with depression outweighs the risk of bleeding events.

Conclusion

SSRIs are widely used as first-line antidepressants all over the world. Most patients with depression need these medications for considerable periods of time. Escitalopram has no effect on the coagulation profile, and although fluoxetine caused a significant increase in the bleeding time after 3 months of treatment, this was not beyond the normal range. Therefore, coagulopathy should not be taken as a contraindication in using SSRIs in patients with hematological disorders and patients undergoing major surgical procedures. Escitalopram and fluoxetine can be used safely for long-term treatment.

Large multicentre trials of antidepressants alone, in combination with NSAIDs and anticoagulants are required to substantiate the findings of this study.

Footnotes

Acknowledgements

Dr Prajakt Barde and Dr Mohini Barde, Shrimohini Centre for Biostatistics are acknowledged for statistical analysis.

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

The author declares no conflicts of interest in preparing this article.

References

American Psychiatric Association (2000) Diagnostic and Statistical Manual of Mental Disorders, 4th edn, text rev. Washington, DC.

Andersohn

Konzen

Bronder

Klimpel

Garbe

(2009). Citalopram induced bleeding due to severe thrombocytopenia. Psychosomatics 50: 297–298.

Andrade

Sandarsh

Chethan

K.B.

Nagesh

K.S.

(2010). Serotonin reuptake inhibitor antidepressants and abnormal bleeding: a review for clinicians and a reconsideration of mechanisms. J Clin Psychiatry 71: 1565–1575.

Baldessani

R.J.

Brunton

L.L.

Lazo

J.S.

Parker

K.L.

(2006) Drug therapy of depression and anxiety disorders. The Pharmacological Basis of Therapeutics, 11th edn. Goodman and Gilman, p. 430.

Dacie

J.V.

Lewis

S.M.

(1975a) Bleeding time – investigation of the hemorrhagic disorders. In: Practical Hematology, 5th edn, pp. 324–325.

Dacie

J.V.

Lewis

S.M.

(1975b). Clotting time – investigation of the hemorrhagic disorders. In: Practical Hematology, 5th edn, pp. 328–329.

Dalton

S.O.

Johenson

Melliemkjaer

Norgard

Sorensen

H.T.

Olsen

J.H.

(2003) Use of selective serotonin reuptake inhibitors and risk of upper gastrointestinal bleeding: a population based cohort study. Arch Intern Med 163: 59–64.

Dalton

S.O.

Sorenson

H.T.

Johenson

(2006) SSRIs and upper gastrointestinal bleeding: what is known and how should it influence prescribing? CNS Drugs 20: 143–151.

Douglas

Smeeth

Irvine

(2011) The use of antidepressants and the risk of haemorrhagic stroke: a nested case control study. Br J Clin Pharmacol 71: 116–120.

10.

Halperin

Reber

(2010) Influence of antidepressants on haemostasis. Eur J Clin Pharmacol 66: 823–829.

11.

Kumar

Saaraswat

Sengupta

S.N.

Mehrotra

(2009) Upper gastrointestinal bleeding in a patient with depression receiving selective serotonin reuptake inhibitor therapy. Indian J Pharmacol 41: 51–53.

12.

Lake

M.B.

Birmaher

Wassick

Mathos

Yelovich

A.K.

(2000) Bleeding and selective serotonin reuptake inhibitors in childhood and adolescence. J Child Adolesc Psychopharmacol 10: 35–38.

13.

Layton

Clark

D.W.

Pearce

G.L.

Shakir

S.A.

(2001) Is there any association between selective serotonin reuptake inhibitors and risk of abnormal bleeding? Results from a cohort study based on prescription event monitoring in England. Eur J Clin Pharmacol 57: 167–176.

14.

Maayan-Metzger

Kuint

Lubetsky

Shenkman

Mazkereth

Kenet

(2006) Maternal selective serotonin reuptake inhibitor intake does not seem to affect neonatal platelet function tests. Acta Haematol 115: 157–161.

15.

Mirsal

Kalyoncu

Pekta

(2002). [Ecchymosis associated with the use of fluoxetine: case report]. Turk Psikiyatri Derg 13: 320–324.

16.

Ramasubbu

(2004) Cerebrovascular effects of selective serotonin reuptake inhibitors: a systematic review. J Clin Psychiatry 65: 1642–1653.

17.

Rang

H.P.

Dale

M.M.

Ritter

J.M.

Moore

P.K.

(2007) Antidepressant drugs. In: Rang and Dales Pharmacology, 7th edn. Amsterdam: Elsevier, p. 556.

18.

Schalekamp

Klungel

O.H.

Souverein

P.C.

de Boer

(2008). Increased bleeding risk with concurrent use of selective serotonin reuptake inhibitors and coumarins. Arch Intern Med 28: 180–185.

19.

Strubel

Birkhofer

Mössmer

Förstl

(2010). SSRI – treatment and bleeding. What risks do we take? Nervenarzt 81: 549–555.

20.

Weinreib

R.M.

Auriacombe

Lynch

K.G.

Lewis

J.D.

(2005) Selective serotonin reuptake inhibitors and the risk of bleeding. Expert Opin Drug Saf 4: 337–344.

21.

Wessinger

Kalpan

Choi

Williams

Lau

Sharp

. (2006) Increased use of selective serotonin reuptake inhibitors in patients admitted with gasterointestinal haemorrhage: a multicentre retrospective analysis. Aliment Pharmacol Ther 23: 937–944.