Should We Routinely Measure and Correct Coagulopathy Prior to Paracentesis in Patients with Liver Disease? A Review of Available Outcome Data

Vol. 29, Issue 1 (March 2024)

	Ana Maria Davila Morales, MD Internal Medicine Resident Rush University Medical Center Chicago, Illinois, USA
	Zoë Post, MD Gastroenterology and Hepatology Fellow Rush University Medical Center Chicago, Illinois, USA
	Costica Aloman, MD Associate Professor, Hepatology & Liver Transplantation Westchester Medical Center Valhalla, New York, USA

Introduction

Paracentesis is a procedure that serves as a diagnostic and therapeutic measure in patients with end stage liver disease (ESLD).¹ Cirrhosis leads to decreased synthesis of pro- and anti-coagulants as well as thrombocytopenia, therefore placing patients at a theoretical risk for post-procedural bleeding. However, paracentesis has been deemed a safe procedure with rarely reported complications when performed using correct technique.² Such complications include infection, bowel perforation and bleeding.² Multiple studies have evaluated the need for coagulation parameters prior to paracentesis to avoid potential bleeding complications, yet no data support cut off values for platelet count or International Normalized Ratio (INR) based on prothrombin time (PT) for which corrective platelet transfusions or fresh frozen plasma (FFP) respectively should be administered pre-procedurally.¹ The aims of this review are 1) to summarize available data on post-procedural outcomes in ESLD patients undergoing paracentesis and 2) to highlight the current guidelines and recommendations regarding routine pre-procedural INR and platelet measurement as well as transfusions to correct coagulopathy in patients with advanced liver disease.

Rates and risk factors of peri-procedural bleeding

Multiple studies have evaluated risk factors and rates of bleeding in ESLD patients undergoing paracentesis. An early study by McVay et al. evaluated 441 paracentesis outcomes and of all patients who did not receive FFP prior to procedure, only one required a transfusion for paracentesis-related bleeding, conferring an event rate of 0.25%. This patient had a normal platelet count and an elevated INR to the same extent as the other 261 patients who underwent paracentesis without complications.³ Patients with markedly elevated serum creatinine levels (6.0 to 14.0 mg/dL) had significantly greater average hemoglobin loss than patients with normal serum creatinine levels, suggesting that these patients deserve close post-procedure observation. In 2005, Pache et al. conducted one of the largest studies addressing this topic. A total of 4,729 patients with ESLD undergoing paracentesis were evaluated over a 10-year period and found severe hemorrhage in nine patients, conferring a rate of 0.2% (Table 1). Bleeding was not associated with INR or platelet count in these patients as only two of the nine patients had significant coagulopathy as evidenced by a platelet count less than 50 or INR greater than two. In fact, bleeding was found to occur in patients with significant pre-existing renal dysfunction, which may suggest uremia as a predisposing factor for bleeding complications.¹ Another large study by Grabau et al. evaluated 1,100 large-volume paracenteses (LVP) performed at a gastroenterology clinic over a seven-year period. No major bleeding events occurred despite 27% of patients having an INR greater than 2.0 and 54% having a platelet count of less than 50,000/μL. Moreover, there were no major bleeding events even in patients with platelet counts being as low as 19,000/μL and INR as high as 8.7.² Kurup et al. evaluated 205 thrombocytopenic patients with platelet counts of less than 50,000 /μL (mean platelet count of 38,400 ±9,300 /μL). Three major bleeding complications requiring red blood cell transfusion were observed in patients with platelet counts of 41,000 to 46,000 /μL, for a complication rate of 0.99%. None of these three patients required any additional procedure or died secondary to bleeding complications and there was no association between platelet count and bleeding complications.⁴ In addition, a study by Webster et al. reviewed the cases of 179 patients with cirrhosis undergoing LVP in an outpatient setting during a one-year period. Of these, four patients (2.2%) developed severe hemorrhagic complications requiring hospital admission and blood product administration.

 

Study author, year	Type of study	Sample size (n)	INR	Platelet count (109/L)	Blood products received	Complication rate
McVay et al., 1991	Prospective cohort study	441	Not evaluated	50-99	Not included in study	0.2%
Pache et al., 2005	Retrospective review	4,729	As high as 8.7 (mean 2.0)	As low as 19 (mean 102)	Not included in study	0.19%
Grabau et al., 2004	Retrospective review	1,100	Range:0.9-8.7	Range:19-341	Not included in study	No bleeding events
Intagliata et al., 2023	Prospective observational study	1,187	Mean 1.7	Mean 128	223 FFP or platelet transfusions (7.8%)	Major bleeding in 2.3% of patient admissions and 0.9% of procedures
Webster et al., 1996	Retrospective review	179	Any level	Any level	Not included in study	2.2%
De Pietri et al., 2016	Randomized control trial	60	INR >0.8	<50	SOC group: 97% TEG group: 17%	SOC group: 3.3% TEG group: 0%
Rowley et al., 2019	Retrospective review	3,116	Mean 1.6	Mean 122	No products	0.19%

Table 1: Overview of studies evaluating bleeding complications in patients with ESLD and coagulopathy. SOC = standard of care, TEG = thromboelastography, INR = international normalized ratio.

Studies have also evaluated post-procedural bleeding in ESLD patients undergoing other nonsurgical procedures such as thoracentesis, lumbar punctures, endoscopic procedures, and biopsies and have found equally low rates of bleeding complications.⁵ A recent multicenter study in 2023 by Intagliata et al. enrolled 1,187 patients undergoing nonsurgical procedures and prospectively monitored them until surgery, transplantation, or 28 days from admission. A total of 93 post-procedural bleeding events were identified, and major bleeding was reported in 2.3% of patient admissions and 0.9% of the procedures. Patients with bleeding were more likely to have nonalcoholic steatohepatitis (NASH) and higher body mass index. Preprocedural INR, platelet level and antithrombotic use were not predictive of bleeding.⁵ This study not only emphasizes the idea that the literature does not support a correlation between degree of coagulopathy and risk of bleeding but also raises the concern for lack of reliable tests or risk scores to assess true bleeding risk in patients with cirrhosis.⁶ Furthermore, a recent expert opinion report published in 2023 also recommends against taking INR into account before performing low-risk procedures in patients with ESLD. However, based only on opinion, the lowest acceptable platelet count was considered to be 30x10⁹/L. This further suggests that platelet count cutoffs lack scientific data and have been largely based on opinion and subjective information.⁷

Risk of bleeding assessment

In cirrhosis, there is a decrease in both pro- and anti-coagulants and there is growing evidence that these patients are “rebalanced” to maintain a tenuous but balanced state of hemostasis. INR level only measures the activity of procoagulants, failing to capture changes in anticoagulants. Therefore, INR may provide a measure of liver function but does not accurately predict bleeding risk in these patients.⁸ Similarly, platelet adherence decreases in cirrhosis however von Willebrand factor (vWF) levels are elevated in proportion to the severity of liver disease. This provides a compensatory change that makes platelet counts an equally ineffective way to predict bleeding risk in patients with advanced liver disease.⁹

The use of INR and platelet count testing to establish arbitrary pre-procedure transfusion guidelines lacks clinical benefit, as evidenced by low procedure-related bleeding rates regardless of coagulopathy or pre-procedural blood product administration. However, post-procedure bleeding events do occur and can be associated with significant morbidity and mortality, underscoring the need to better identify the patients at increased risk. Thromboelastography (TEG) is a measure of blood coagulation efficiency and studies have evaluated its utility in the assessment and reversal of coagulopathy among ESLD patients as compared to standard coagulation testing.¹⁰ A systematic review and meta-analysis by Kovalic et al. found that there was a significant reduction in number of patients requiring coagulopathy-corrective transfusions when using TEG. In addition, this method was found to improve overall number of patients exposed to blood product transfusions, quantity of transfusions and bleeding events. To better define the patient population that could benefit from pre-procedural correction of coagulopathy, De Pietri et al. performed a randomized controlled trial in 2015 aiming to define the efficacy and safety of TEG in guiding the use of FFP or platelet transfusion before invasive procedures in patients with liver cirrhosis and impaired coagulation parameters. Enrollment criteria included INR greater than 1.8 and/or platelet count less than 50,000/μL. 60 patients were randomly allocated to TEG-guided transfusion strategy or standard of care (SOC). The TEG group would receive FFP if the reaction time was >40 min and/or platelet transfusion if maximum amplitude was <30 mm while the SOC group received transfusions per hospital guidelines. In the SOC group, all but one patient received blood product transfusions versus five in the TEG group (97% vs 17%), suggesting that TEG-guided transfusion strategies lead to significantly lower use of blood products compared to SOC without an increase in bleeding complications.¹¹

In addition, fibrinogen levels have recently been viewed as potentially more meaningful than INR and platelet levels as a measure of bleeding risk.¹² The current American Gastroenterological Association (AGA) Clinical Practice Guideline from 2019 recommends a fibrinogen transfusion threshold of >120 mg/dL, however, this is only for active bleeding or high-risk procedures. There is no guidance on transfusion thresholds for low-risk procedures including LVP. While data on fibrinogen level and procedural outcomes is limited, a small retrospective study by Lin et al. demonstrated that a low fibrinogen level could independently predict bleeding events in acute on chronic liver failure (ACLF) patients with a MELD score of at least 25.¹³ Certainly, more studies are needed to develop clinically relevant recommendations in terms of pre-procedure fibrinogen testing and transfusion thresholds. When fibrinogen repletion is desired, cryoprecipitate can be used and has less impact on volume expansion and portal hypertension as it contains smaller volume than FFP.¹² It has also been suggested that lower volume factor replacement such as prothrombin complex complexes (PCC) and recombinant factor VIIa are more effective than FFP in decreasing INR values in ESLD patients and do not carry the risk of volume overload.¹⁴

Current Guidelines

Based on the aforementioned studies, guidelines have been established by multiple professional societies to provide guidance on how to safely manage ESLD patients undergoing paracentesis (Table 2).

According to the Society of Hospital Medicine, studies have widely demonstrated that neither INR nor platelet count accurately predict bleeding risk in this population and there is lack of benefit from routine measurement and administration of corrective products.⁶ The American Gastroenterological Association (AGA) also suggests against the use of extensive preprocedural testing such as PT/INR or platelet count, and the Society of Interventional Radiology (SIR) recommends against using an INR threshold for low-risk procedures such as paracentesis but may be considered for patients receiving warfarin or heparin products.^{15, 16} In patients with disseminated intravascular coagulation (DIC) and fibrinolysis, coagulation parameters should always be obtained to move forward with paracentesis.¹

Guideline	Pre-procedural INR/Platelet testing	Pre-procedure transfusion thresholds
SHM	None
AGA	None
SIR	None	Consider PLT goal >20,000
AASLD		No prophylaxis

Table 2. Overview of (inter)national guideline recommendations on pre-procedural testing and transfusion thresholds for low-risk procedures including paracentesis. SHM = Society of Hospital Medicine; AGA = American Gastroenterological Association; SIR = Society of Interventional Radiology; AASLD = American Association for the Study of Liver Diseases.

Currently, the American Association for the Study of Liver Diseases (AASLD) recommends against the routine prophylactic use of FFP or platelets before paracentesis.¹³ However, the SIR recommends that platelet transfusions should be considered for low-bleeding risk procedures when platelet count is <20,000/μL.¹⁶

Impact of unnecessary correction of coagulopathy in ESLD patients

Despite all the recommendations by professional societies as outlined above, there are multiple reports regarding the negative impact of unnecessary correction of coagulopathy in patients with ESLD. Barnhill et al. found that at a United States (US) tertiary care center in Washington D.C., out of 177 patients undergoing LVP from 2011 to 2015, 22.2% received prophylactic FFP and 17.3% received prophylactic platelet transfusions before paracentesis, demonstrating the large number of patients still undergoing seemingly unnecessary product transfusions.¹⁷ In the presence of substantial evidence that these parameters do not accurately predict bleeding risk, unnecessary transfusions are contributing to procedural delays, bacterial contamination, and transfusion-associated adverse events such as anaphylaxis, circulatory overload, and lung injuries.⁶ Specifically in patients with decompensated liver cirrhosis, volume expansion from transfusions may contribute to worsening portal hypertension and increase risk of variceal bleeding.⁶ Unnecessary transfusions are also a burden to healthcare-related costs. Rowley et al. aimed to evaluate the rate and risk factors for hemorrhage in patients undergoing ultrasound-guided paracentesis without correction of coagulopathy. Over a two-year period, 3116 paracentesis procedures were performed in both outpatient and inpatient settings. Mean INR was 1.6 and 437 patients (14%) had an INR >2. Mean platelet count was 122,000/μL and 368 patients (12%) had a platelet count <50,000/μL. Without correction of coagulopathy, significant post-paracentesis hemorrhage occurred in five (0.19%) patients, and only one patient required an angiogram with embolization. Transfusion of 1,125 units of FFP and 366 units of platelets were avoided, for a transfusion-associated cost savings of $816,000 over two years.¹⁸

Indications for pre-procedural coagulopathy correction

There are a few exceptions where pre-procedural transfusions to correct coagulopathy in patients with ESLD are indicated, including patients with DIC and hyperfibrinolysis.⁶ If patients lack signs and symptoms of these two clinical scenarios, it is recommended that providers avoid routine measurement of INR and platelet count in preparation for paracentesis. Moreover, providers should avoid routine transfusion of FFP and platelets prior to paracentesis even in the presence of abnormal coagulation parameters.

Conclusion

There is substantial evidence in the literature demonstrating that neither INR nor platelet counts accurately predict bleeding risks in cirrhotic patients undergoing paracentesis. Providers should not routinely measure coagulation parameters in cirrhotic patients in preparation of a low-risk procedure such as paracentesis and should avoid routine platelet and FFP transfusions. Exceptions include patients with DIC or hyperfibrinolysis. Implementing these measures could result in fewer procedural delays, bacterial contamination, healthcare-related costs, transfusion-associated reactions, and volume expansion leading to worsening portal hypertension.

REFERENCES

1. Pache, I. and Bilodeau, M. (2005), Severe haemorrhage following abdominal paracentesis for ascites in patients with liver disease. Alimentary Pharmacology & Therapeutics, 21: 525-529. https://doi.org/10.1111/j.1365-2036.2005.02387.x

2. Grabau, C.M., Crago, S.F., Hoff, L.K., Simon, J.A., Melton, C.A., Ott, B.J. and Kamath, P.S. (2004), Performance standards for therapeutic abdominal paracentesis. Hepatology, 40: 484-488. https://doi.org/10.1002/hep.20317

3. McVay, P.A. and Toy, P.T.C.Y. (1991), Lack of increased bleeding after paracentesis and thoracentesis in patients with mild coagulation abnormalities. Transfusion, 31: 164-171. https://doi.org/10.1046/j.1537-2995.1991.31291142949.x

4. Kurup, A.N., Lekah, A., Reardon, S.T., Schmit, G.D., McDonald, J.S., Carter, R.E., Kamath, P.S., Callstrom, M.R. and Atwell, T.D. (2015), Bleeding Rate for Ultrasound-Guided Paracentesis in Thrombocytopenic Patients. Journal of Ultrasound in Medicine, 34: 1833-1838. https://doi.org/10.7863/ultra.14.10034

5. Intagliata NM, Rahimi RS, Higuera-de-la-Tijera F, Simonetto DA, Farias AQ, Mazo DF, Boike JR, Stine JG, Serper M, Pereira G, Mattos AZ, Marciano S, Davis JPE, Benitez C, Chadha R, Méndez-Sánchez N, deLemos AS, Mohanty A, Dirchwolf M, Fortune BE, Northup PG, Patrie JT, Caldwell SH. Procedural-Related Bleeding in Hospitalized Patients With Liver Disease (PROC-BLeeD): An International, Prospective, Multicenter Observational Study. Gastroenterology. 2023 Sep;165(3):717-732. doi: 10.1053/j.gastro.2023.05.046. Epub 2023 Jun 2. PMID: 37271290.

6. Crowe, Byron, et al. “Things We Do for No ReasonTM: Routine Correction of Elevated INR and Thrombocytopenia Prior to Paracentesis in Patients with Cirrhosis.” Journal of Hospital Medicine, Society of Hospital Medicine, Sept. 2020, cdn.mdedge.com/files/s3fs-public/issues/articles/lessing08150923e.pdf. 

7. Riescher-Tuczkiewicz, A., Caldwell, S. H., Kamath, P. S., Villa, E., & Rautou, P.-E. (2023). Expert opinion on bleeding risk from invasive procedures in cirrhosis. JHEP Reports, 100986. https://doi.org/10.1016/j.jhepr.2023.100986 

8. Tripodi, A., Caldwell, S.H., Hoffman, M., Trotter, J.F. and Sanyal, A.J. (2007), Review article: the prothrombin time test as a measure of bleeding risk and prognosis in liver disease. Alimentary Pharmacology & Therapeutics, 26: 141-148. https://doi.org/10.1111/j.1365-2036.2007.03369.x

9. Northup, P. G., & Caldwell, S. H. (2013). Coagulation in liver disease: A guide for the clinician. Perspectives in Clinical Gastroenterology and Hepatology, 11: 1064-1074. https://doi.org/10.1016/j.cgh.2013.02.026 

10. Kovalic AJ, Khan MA, Malaver D, Whitson MJ, Teperman LW, Bernstein DE, Singal A, Satapathy SK. Thromboelastography versus standard coagulation testing in the assessment and reversal of coagulopathy among cirrhotics: a systematic review and meta-analysis. Eur J Gastroenterol Hepatol. 2020 Mar;32(3):291-302. doi: 10.1097/MEG.0000000000001588. PMID: 32012141.

11. De Pietri, L., Bianchini, M., Montalti, R., De Maria, N., Di Maira, T., Begliomini, B., Gerunda, G.E., di Benedetto, F., Garcia-Tsao, G. and Villa, E. (2016), Thrombelastography-guided blood product use before invasive procedures in cirrhosis with severe coagulopathy: A randomized, controlled trial. Hepatology, 63: 566-573. https://doi.org/10.1002/hep.28148

12. Liu P, Hum J, Jou J, Scanlan RM, Shatzel J. Transfusion strategies in patients with cirrhosis. Eur J Haematol. 2020 Jan;104(1):15-25. doi: 10.1111/ejh.13342. Epub 2019 Nov 19. PMID: 31661175; PMCID: PMC7023893.

13. Jacqueline G. O’Leary, Charles S. Greenberg, Heather M. Patton, Stephen H. Caldwell. AGA Clinical Practice Update: Coagulation in Cirrhosis, Gastroenterology. Volume 157, Issue 1, Pages 34-43.e1. (2019). https://doi.org/10.1053/j.gastro.2019.03.070.

14. Lin S, Wang M, Zhu Y, et al. Hemorrhagic Complications Following Abdominal Paracentesis in Acute on Chronic Liver Failure: A Propensity Score Analysis. Medicine (Baltimore). 2015;94(49):e2225. doi:10.1097/MD.0000000000002225

15. Patrick G. Northup, Ton Lisman, Lara N. Roberts. Treatment of bleeding in patients with liver disease. Journal of Thrombosis and Haemostasis. Volume 19, Issue 7, Pages 1644-1652 (2021). ISSN 1538-7836. https://doi.org/10.1111/jth.15364.

16. O'Leary JG, Greenberg CS, Patton HM, Caldwell SH. AGA Clinical Practice Update: Coagulation in Cirrhosis. Gastroenterology. 2019 Jul;157(1):34-43.e1. doi: 10.1053/j.gastro.2019.03.070. Epub 2019 Apr 12. PMID: 30986390.

17. Biolato M, Vitale F, Galasso T, Gasbarrini A, Grieco A. Minimum platelet count threshold before invasive procedures in cirrhosis: Evolution of the guidelines. World J Gastrointest Surg. 2023 Feb 27;15(2):127-141. doi: 10.4240/wjgs.v15.i2.127. PMID: 36896308; PMCID: PMC9988645.

18. Barnhill, M., Lee, A., and Montero, A (2017). Adherence Rates to Recommended Guidelines for Paracentesis in Cirrhotic Patients at a Tertiary Care Center and Associated Complications. The American Journal of Gastroenterology, 112: p504. https://journals.lww.com/ajg/fulltext/2017/10001/adherence_rates_to_recommended_guidelines_for.899.aspx

19. Rowley MW, Agarwal S, Seetharam AB, Hirsch KS. Real-Time Ultrasound-Guided Paracentesis by Radiologists: Near Zero Risk of Hemorrhage without Correction of Coagulopathy. J Vasc Interv Radiol. 2019 Feb;30(2):259-264. doi: 10.1016/j.jvir.2018.11.001. PMID: 30717961.