- Email: [email protected]
Thrombocytopenia and Neurosurgery: A Literature Review
Accepted Manuscript Thrombocytopenia and Neurosurgery: A Literature Review Daphne Li, M.D., Taylor Glor, B.Sc., G. Alexander Jones, M.D. PII:
S1878-8750(17)30989-0
DOI:
10.1016/j.wneu.2017.06.097
Reference:
WNEU 5968
To appear in:
World Neurosurgery
Received Date: 8 March 2017 Revised Date:
12 June 2017
Accepted Date: 14 June 2017
Please cite this article as: Li D, Glor T, Jones GA, Thrombocytopenia and Neurosurgery: A Literature Review, World Neurosurgery (2017), doi: 10.1016/j.wneu.2017.06.097. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Thrombocytopenia and Neurosurgery: A Literature Review
EP
TE D
Daphne Li, M.D.1, Taylor Glor, B.Sc.2, G. Alexander Jones, M.D.1
Department of Neurological Surgery, Loyola University School of Medicine, Maywood, IL
2
Loyola University Stritch School of Medicine, Maywood, IL
AC C
1
ACCEPTED MANUSCRIPT
Corresponding Author: G. Alexander Jones, M.D., F.A.A.N.S. Assistant Professor
RI PT
Loyola University School of Medicine Department of Neurological Surgery c/o Ms. Albana Isaj
SC
Maguire Building 105, Room 1900
Maywood, IL 60153 Phone: 708-216-9032 Fax: 708-216-4948
TE D
Email: [email protected]
M AN U
2160 South First Avenue
AC C
EP
Keywords: thrombocytopenia, neurosurgery, perioperative hemorrhage risk, platelet transfusion
ACCEPTED MANUSCRIPT
Abstract The absence of evidence-based guidelines for platelet transfusion surrounding invasive
RI PT
neurosurgical procedures leads to uncertainty in management. Multiple studies have illuminated this lack of high quality data, and subsequent reliance on expert opinion. The generally accepted threshold for platelet transfusion has hovered around 100,000/µl. We have conducted a review of available clinical literature, in order to identify any evidence that may support or refute these
SC
general guidelines, in an attempt to clarify the need for platelet transfusion in the patient
requiring neurosurgical intervention. The available evidence is sparse and of low quality, but suggests that a platelet count below 100,000/µl is associated with increased risk of hemorrhagic
M AN U
complications. Additionally, the acuity of thrombocytopenia, magnitude of decrease in platelet count and responsiveness to platelet transfusions impact the risk of neurosurgical intervention, and should be taken into account when evaluating a patient’s surgical candidacy. Higher quality, prospective studies on the subject are unlikely, given a general lack of clinical equipoise on the
AC C
EP
TE D
subject, and the ethical concerns such a study would present.
ACCEPTED MANUSCRIPT
Introduction Neurosurgical procedures require impeccable hemostasis, as even a small amount of hemorrhage
RI PT
may cause catastrophic damage. Therefore, especially prior to elective neurosurgical procedures, hemostatic parameters are typically assessed. Particular attention is paid to any indicator that may suggest an increased risk for bleeding intra- or postoperatively. Thrombocytopenia is
known to be associated with an increased risk of hemorrhage, and evidence-based guidelines for
SC
platelet transfusion thresholds have been established for non-neuraxial procedures.1-3 For neurosurgical procedures, the generally accepted recommendation is that a platelet count
M AN U
threshold of 100,000/µl should be used, to minimize the risk of hemorrhagic complications. While this practice has been handed down through several generations of neurosurgical training, this is based largely on expert opinion, and there is a paucity of evidence to support this guideline concerning neurosurgical patients. Thus, we have performed a review of the literature, for the purpose of evaluating the available evidence regarding platelet transfusion thresholds in neurosurgical patients. In order to broaden the scope of our review, we have also chosen to
patients. Methods
TE D
include articles evaluating thrombocytopenia, and its associated risks, in traumatic brain injury
EP
Two computerized searches of the MEDLINE database (Pubmed) from January, 1966 to November, 2016 were conducted: (1) keyword search using “neurosurgery AND platelets,” and
AC C
“neurosurgical procedures AND platelets”; (2) keyword search using similar combinations of “platelets,” “thrombocytopenia,” “intracranial hemorrhage,” "subdural hematoma", and “trauma”. Articles were selected for inclusion based on relevance. Only articles published in the English language were included. In addition, any relevant articles referenced in the originally selected articles were considered for inclusion as well. Articles were excluded if there was no documentation of preoperative thrombocytopenia.
ACCEPTED MANUSCRIPT
Results In general, the articles fell into one of four categories: neurosurgical case reviews; NSQIP
RI PT
database studies; studies of thrombocytopenia in head trauma; and professional society guidelines based on expert opinion. Case Reviews
SC
Two retrospective case reviews were found in the literature review regarding thrombocytopenia and perioperative hemorrhage in neurosurgical patients. Schramm et al., reported on the routine testing of coagulation studies in preoperative neurosurgical patients and only found an
M AN U
association between prolonged APTT and postoperative bleeding.4 No clear association was found for thrombocytopenia, but these authors did not clearly define what threshold was used in order to classify a patient as thrombocytopenic.
Chan et al., conducted a retrospective case review in 1989, which has since been frequently cited, showing a significant association between platelet counts below 100,000/µl with
TE D
postoperative hematoma.5 A failure to respond to transfusion was associated with an increased risk of postoperative hematoma in thrombocytopenic patients. Of note, Chan et al also found that a precipitous drop in platelets in the perioperative period was highly significant: the mean percentage decrease in platelets for patients with hemorrhage was 58.5%, compared with 30.7%
EP
for those without. Greater acute decreases in platelet count, even if the absolute value remained
AC C
above 100,000/µl, were associated with increased risk of postoperative hematoma. Abdelfatah performed a retrospective review of the management of 41 patients with chronic subdural hematomas and intractable thrombocytopenia – defined as platelet count not reaching 100,000/µl after platelet transfusions (18 units of random donor pooled platelets preoperatively, 6 units intaoperatively and 6 units postoperatively).6 Patients had platelet counts ranging from 23,000 to 62,000/µl on admission; the last preoperative platelet count ranged from 51,000 to 79,000/µl. Some had coagulopathies and/or anemia that was also corrected with FFP or packed red blood cell transfusions. The patients in this review were all successfully treated with burr hole evacuations without recurrent SDH. The outcomes of these chronically thrombocytopenia
ACCEPTED MANUSCRIPT
patients suggests that the acuity of a patient's thrombocytopenia may play a role, and that chronically thrombocytopenic patients may be successfully treated with a platelet count below 100,000/µl.
RI PT
NSQIP Three NSQIP database reviews published recently with the goals of evaluating hemostatic
screening and/or thrombocytopenia in neurosurgical patients. Dasenbrock et al., conducted an analysis of NSQIP data with a focus on patients undergoing craniotomy for tumor resection.7
SC
They found that the 30-day mortality rate was significantly higher for patients with moderate (100-124,000/µl), severe (75-99,000/µl) or very severe (<74,000/µl) thrombocytopenia. As well,
M AN U
the rate of unplanned reoperation, especially for postoperative intracranial hemorrhage, was significantly higher for patients with thrombocytopenia classified as severe or very severe. Bronheim et al., conducted an analysis of the NSQIP database with a focus on coagulation parameters in patients undergoing posterior lumbar fusion, one of the most common surgical procedures that routinely requires transfusion.8 Their analyses suggest that a platelet count of <150,000/µl was a significant independent risk factor for transfusion, organ space surgical site
TE D
infection, acute renal failure, reoperation, prolonged ventilation (>48 hours), sepsis, and death. Another NSQIP database analysis, conducted by Seicean et al., focused on the use and utility of preoperative hemostatic screening in adult neurosurgical patients.9 They found a significant
EP
association between abnormal hemostatic parameters and blood transfusions, return to the operating room, and mortality. However, when analyzed individually, platelet count had low
AC C
sensitivity and specificity for predicting any of the aforementioned complications. They did find that an abnormal platelet count was associated with increased intraoperative transfusions, but did not find any significant association with increased bleeding risk. They concluded that preoperative hemostatic screening would be most cost effective when limited to patients with a history of abnormal hemostasis. Trauma Due to the limited evidence found regarding platelet counts and complications in patients undergoing elective neurosurgical procedures, we extended our review of the literature to include
ACCEPTED MANUSCRIPT
investigations on coagulopathies in patients with traumatic injuries involving the central nervous system. Coagulopathy is a well-established predictor of mortality in patients with traumatic brain injury, but investigation on the specific impact of platelets remains scarce.
RI PT
The most notable article in this category is a Cochrane Review conducted by Yuan et al., published in the Journal of Neurotrauma.10 They performed meta-analyses of 19 studies, with a focus on the effects of coagulopathy in patients with traumatic brain injury (TBI), and its effect on progressive hemorrhagic injury (PHI). A small number of studies were included in two meta-
SC
analyses studying the association of thrombocytopenia with the development of TBI; neither was significant. Two published studies failed to show a difference in platelet count when comparing TBI and non-TBI patients. Three studies showed no difference in platelet count in a comparison
M AN U
of isolated TBI patients versus TBI patients with other traumatic injuries. They reviewed fifteen studies comparing coagulopathy in patients with PHI versus those without PHI. They found that patients with evidence of PHI did have a lower platelet count, usually <100,000/µl, and an INR of >1.2. In addition, the authors of this review suggested that the use of thromboelastography for clot strength analysis may be used to more effectively guide transfusion practice, rather than
TE D
simply following quantitative coagulation parameters.
Schnuriger et al., performed a focused retrospective chart review of 310 blunt head trauma patients in order to investigate the impact of platelets on the progression of hemorrhagic injury.11 They found that patients with a platelet count of <100,000/µl had a ninefold increased adjusted
EP
risk of mortality, independent of other coagulation parameters (INR and aPTT). In addition, a platelet count of <175,000/µl was found to be a significant predictor of progression of
AC C
hemorrhagic injury. However, in spite of the association of thrombocytopenia with increased mortality, the platelet count did have a poor sensitivity (11%) for predicting worsening intracranial hemorrhage. Therefore, the applicability of this finding remains uncertain. Bullock et al., reviewed the craniotomy data for 850 patients that underwent evacuation of a post-traumatic intracranial hematoma and identified 59 patients who required a second operation for removal of a second intracranial lesion.12 Coagulopathy was identified as a risk factor associated with post-craniotomy hematoma. More specifically, thrombocytopenia was detected in 19 patients (32%), with platelet counts ranging from 26,000 to 129,000/µl. Serial platelet
ACCEPTED MANUSCRIPT
counts were noted to be lowest on the day after the first craniotomy and in three patients, platelet transfusions were given before their second operation. Thrombocytopenia was not found to be associated with a significantly worse outcome in this case review.
RI PT
Multiple smaller sample size case reviews were published that focused on coagulopathy and the outcome of patients with severe head injury. However, amidst the multiple parameters evaluated for coagulopathy, only limited focus has been placed on platelet count. Gupta et al., did a
prospective case review of 50 patients and did find that thrombocytopenia – defined as platelet count of less than 100,000 – as an independent risk factor for worse clinical outcome in patients
SC
with moderate head injury, but not those with severe head injury.13 Kuo et al., evaluated a group of 61 head injury patients by a coagulopathy score and provided evidence this was a better
M AN U
predictor of mortality than radiographic findings of brain compression.14 However, the mean platelet counts were within the normal range, and only two of 61 patients were thrombocytopenic. Salehpour et al., found that the median serum platelet count was significantly lower in patients with severe isolated head traumas (GCS < 8) who did not survive, although it was >200,000/µl in both groups.15 Konig et al., studied coagulopathy in patients with chronic subdural hematomas and did find a significant difference in patients with recurrent (mean
AC C
Guidelines
EP
subdural hematomas.16
TE D
platelets count 177,000/µl) and non-recurrent (mean platelet count of 226,000/µl) chronic
The American Society of Hematology last published guidelines on platelet transfusion in 2007, the Journal of Hospital Medicine published periprocedural transfusion guidelines in 2013, and the American Association of Blood Banks (AABB) cited various recommendations regarding platelet transfusion thresholds in 2015.1-3 The consensus is that there is poor or low quality data to suggest a platelet transfusion threshold for neuraxial procedures or patients with intracranial hemorrhage. However, due to the potential for significant adverse outcomes, it is expert opinion that platelets should be maintained above 100,000/µl in patients with intracerebral bleeding. More specifically, the AABB recommends prophylactic platelet transfusion for patients
ACCEPTED MANUSCRIPT
undergoing elective diagnostic lumbar punctures with plalelet counts less than 50,000/µl.1 The AABB cannot recommend for or against platelet transfusion for patients with antiplatelet therapy who had intracranial hemorrhage. For patients undergoing nonneuraxial surgery or with nonneuraxial bleeding prophylactic platelet transfusions are recommended if the platelet count is
RI PT
less than 50,000/µl. Discussion
A platelet transfusion threshold of 100,000/µl in patients requiring elective neurosurgical
SC
procedures and in patients with acute neuraxial hemorrhages is a widely accepted standard of practice. However, a review of the literature reveals a paucity of evidence surrounding this topic.
M AN U
The majority of the data consist of retrospective cohort studies or case reviews. In addition, official published guidelines either cite expert opinion or state that they are unable to make formal recommendations regarding the safety of neurosurgical procedures in the setting of thrombocytopenia.1-3
It is understandable that there is a lack of randomized control trials, or even prospective studies, surrounding a platelet transfusion thresholds in patients undergoing elective neurosurgical
TE D
procedures. The risks of intra-operative and post-operative hemorrhage in the confines of the central nervous system carry a much too high morbidity and mortality to justify any such investigation. Thus, this literature review has also explored the trauma literature in an effort to
EP
bolster any evidence for a platelet threshold of 100,000/µl. Some findings in the trauma literature do indicate that thrombocytopenia, with a platelet count of less than 100,000/µl, was associated with increased risk of PHI in traumatic brain injury patients.10 Thrombocytopenia less than
AC C
100,000/µl was also found to be associated with a higher mortality rate.11 The evidence on thrombocytopenia as an independent risk factor alone is still limited as the studies on isolated neuro traumas are few and many polytrauma patients are also at risk for other coagulopathies. 1215
Most articles evaluating the risks of hemorrhage or progressive hemorrhage in the neurosurgical patient cite the retrospective case series conducted by Chan, et al in 1989. The conclusions drawn from this investigation seem to have resonated the most with expert opinion and current practice. Although a platelet transfusion threshold of less than 100,000/µl may be observed as
ACCEPTED MANUSCRIPT
general practice, there are other factors that should be taken into account – platelet transfusion responsiveness; the chronicity of a patient's thrombocytopenia; the acuity and severity of the drop in platelet count as well as the urgency of neurosurgical intervention. In addition, the literature surrounding evaluating patients with coagulopathies is trending towards alternative
RI PT
assessments of coagulability rather than relying purely on absolute platelet count. Elective
neurosurgical oncology patients often suffer from chronic thrombocytopenia due to bone marrow suppression or other etiologies. In these settings, tools, such as thromboelastography, may be
SC
useful in patients with chronic thrombocytopenia. Conclusion
M AN U
It has been standard practice to maintain a patient's platelet count above 100,000/µl in elective neurosurgical patients during the perioperative period. Guidelines regarding platelet transfusion thresholds in the neurosurgical perioperative period are founded on weaker clinical evidence and expert opinion. However, the recommendation of a platelet count greater than 100,000/µl prior to elective neurosurgical procedures is unlikely to be further examined due to the mortality and morbidity associated with increased risk of hemorrhage in the neurosurgical patient. Further
TE D
investigation may be done as to risks of neurosurgical intervention in patients with chronic thrombocytopenia and or those unresponsive to platelet transfusion as well as other tools that
AC C
EP
may be used to assess coagulopathy and bleeding risk in these patients.
References 1. Kaufman RM, Djulbegovic B, Gernsheimer T, Kleinman S, Tinmouth AT, Capocelli KE, Cipolle MD, Cohn CS, Fung MK, Grossman BJ, Mintz PD, O'Malley BA, Sesok-Pizzini DA, Shander A, Stack GE, Webert KE, Weinstein R, Welch BG, Whitman GJ, Wong
ACCEPTED MANUSCRIPT
EC, Tobian AA. Platelet Transfusion: a clinical practice guideline from the AABB. Ann Intern Med. 2015;162:205-213. 2. Slichter, SJ. Evidence-based platelet transfusion guidelines. Hematology Am Soc Hematol Educ Program. 2007:172-178.
blood products. J Hosp Med. 2013 Nov;8(11):647-52.
RI PT
3. Hogshire LC, Patel MS, Rivera E, Carson JL. Evidence review: periprocedural use of
4. Schramm B, Leslie K, Myles PS, Hogan CJ. Coagulation studies in preoperative neurosurgical patients. Anaesth Intensive Care 2001; 29:388-392.
SC
5. Chan KH, Mann KS, Chan TK. The significance of thrombocytopenia in the development of postop intracranial hematoma. J Neurosurg 71:38-41, 1989.
M AN U
6. Abdelfatah, M. Management of chronic subdural hematoma in patients with intractable
thrombocytopenia. Turk Neurosurg; 2016.
7. Dasenbrock HH, Devine CA, Liu KX, Gormley WB, Claus EB, Smith TR, Dunn IF. Thrombocytopenia and craniotomy for tumor: a NSQIP analysis. Cancer 2016;122:170817.
8. Bronheim RS, Oermann EK, Cho SK, Caridi JM. Coagulation profile as a risk factor for
TE D
30-day morbidity and mortality following posterior lumbar fusion. Spine 2016. 9. Seicean A, Schiltz NK, Seicean S, Alan N, Neuhauser D, Weil RJ. Use and utility of preoperative hemostatic screening and patient history in adult neurosurgical patients. J Neurosurg 116:1097-1105, 2012.
EP
10. Yuan Q, Sun YR, Wu X, Yu J, Li ZQ, Du ZY, Wu XH, Zhou LF, Hu J. Coagulopathy in Traumatic Brain Injury and Its Correlation with Progressive Hemorrhagic Injury: A
AC C
Systematic Review and Meta-Analysis. J Neurotrauma. 2016 Jul 15;33(14):1279-91. 11. Schnüriger B1, Inaba K, Abdelsayed GA, Lustenberger T, Eberle BM, Barmparas G, Talving P, Demetriades D. The impact of platelets on the progression of traumatic intracranial hemorrhage. J Trauma. 2010 Apr;68(4):881-5.
12. Bullock R, Hannemann CO, Murray L, Teasdale GM. Recurrent hematomas following craniotomy for traumatic intracranial mass. J Neurosurg 72:9-14, 1990. 13. Gupta G, Wadhwa C, Garg R, Dhaiya RS, Kaushal RK. Impact of Coagulation Profile on Outcome of Head Injury. J Clin Diagn Res. 2016 Jan;10(1):PC04-6. 14. Kuo JR1, Chou TJ, Chio CC. Coagulopathy as a parameter to predict the outcome in head
ACCEPTED MANUSCRIPT
injury patients--analysis of 61 cases. J Clin Neurosci. 2004 Sep;11(7):710-4. 15. Salehpour F1, Bazzazi AM, Porhomayon J, Nader ND. Correlation between coagulopathy and outcome in severe head trauma in neurointensive care and trauma units. J Crit Care. 2011 Aug;26(4):352-6.
RI PT
16. König SA1, Schick U, Döhnert J, Goldammer A, Vitzthum HE. Coagulopathy and outcome in patients with chronic subdural haematoma. Acta Neurol Scand. 2003
AC C
EP
TE D
M AN U
SC
Feb;107(2):110-6.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
NSQIP: National Surgical Quality Improvement Program TBI: traumatic brain injury PHI: progressive hemorrhagic injury INR: international normalized ratio aPTT: activated partial thromboplastin time AABB: American Association of Blood Banks
RI PT
ABBREVIATIONS
ACCEPTED MANUSCRIPT
Disclosure
This research did not receive any specific grant from funding agencies in the public,
RI PT
commercial, or not-for-profit sectors. The authors have no financial conflict of interst to
AC C
EP
TE D
M AN U
SC
disclose.
S1878-8750(17)30989-0
DOI:
10.1016/j.wneu.2017.06.097
Reference:
WNEU 5968
To appear in:
World Neurosurgery
Received Date: 8 March 2017 Revised Date:
12 June 2017
Accepted Date: 14 June 2017
Please cite this article as: Li D, Glor T, Jones GA, Thrombocytopenia and Neurosurgery: A Literature Review, World Neurosurgery (2017), doi: 10.1016/j.wneu.2017.06.097. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Thrombocytopenia and Neurosurgery: A Literature Review
EP
TE D
Daphne Li, M.D.1, Taylor Glor, B.Sc.2, G. Alexander Jones, M.D.1
Department of Neurological Surgery, Loyola University School of Medicine, Maywood, IL
2
Loyola University Stritch School of Medicine, Maywood, IL
AC C
1
ACCEPTED MANUSCRIPT
Corresponding Author: G. Alexander Jones, M.D., F.A.A.N.S. Assistant Professor
RI PT
Loyola University School of Medicine Department of Neurological Surgery c/o Ms. Albana Isaj
SC
Maguire Building 105, Room 1900
Maywood, IL 60153 Phone: 708-216-9032 Fax: 708-216-4948
TE D
Email: [email protected]
M AN U
2160 South First Avenue
AC C
EP
Keywords: thrombocytopenia, neurosurgery, perioperative hemorrhage risk, platelet transfusion
ACCEPTED MANUSCRIPT
Abstract The absence of evidence-based guidelines for platelet transfusion surrounding invasive
RI PT
neurosurgical procedures leads to uncertainty in management. Multiple studies have illuminated this lack of high quality data, and subsequent reliance on expert opinion. The generally accepted threshold for platelet transfusion has hovered around 100,000/µl. We have conducted a review of available clinical literature, in order to identify any evidence that may support or refute these
SC
general guidelines, in an attempt to clarify the need for platelet transfusion in the patient
requiring neurosurgical intervention. The available evidence is sparse and of low quality, but suggests that a platelet count below 100,000/µl is associated with increased risk of hemorrhagic
M AN U
complications. Additionally, the acuity of thrombocytopenia, magnitude of decrease in platelet count and responsiveness to platelet transfusions impact the risk of neurosurgical intervention, and should be taken into account when evaluating a patient’s surgical candidacy. Higher quality, prospective studies on the subject are unlikely, given a general lack of clinical equipoise on the
AC C
EP
TE D
subject, and the ethical concerns such a study would present.
ACCEPTED MANUSCRIPT
Introduction Neurosurgical procedures require impeccable hemostasis, as even a small amount of hemorrhage
RI PT
may cause catastrophic damage. Therefore, especially prior to elective neurosurgical procedures, hemostatic parameters are typically assessed. Particular attention is paid to any indicator that may suggest an increased risk for bleeding intra- or postoperatively. Thrombocytopenia is
known to be associated with an increased risk of hemorrhage, and evidence-based guidelines for
SC
platelet transfusion thresholds have been established for non-neuraxial procedures.1-3 For neurosurgical procedures, the generally accepted recommendation is that a platelet count
M AN U
threshold of 100,000/µl should be used, to minimize the risk of hemorrhagic complications. While this practice has been handed down through several generations of neurosurgical training, this is based largely on expert opinion, and there is a paucity of evidence to support this guideline concerning neurosurgical patients. Thus, we have performed a review of the literature, for the purpose of evaluating the available evidence regarding platelet transfusion thresholds in neurosurgical patients. In order to broaden the scope of our review, we have also chosen to
patients. Methods
TE D
include articles evaluating thrombocytopenia, and its associated risks, in traumatic brain injury
EP
Two computerized searches of the MEDLINE database (Pubmed) from January, 1966 to November, 2016 were conducted: (1) keyword search using “neurosurgery AND platelets,” and
AC C
“neurosurgical procedures AND platelets”; (2) keyword search using similar combinations of “platelets,” “thrombocytopenia,” “intracranial hemorrhage,” "subdural hematoma", and “trauma”. Articles were selected for inclusion based on relevance. Only articles published in the English language were included. In addition, any relevant articles referenced in the originally selected articles were considered for inclusion as well. Articles were excluded if there was no documentation of preoperative thrombocytopenia.
ACCEPTED MANUSCRIPT
Results In general, the articles fell into one of four categories: neurosurgical case reviews; NSQIP
RI PT
database studies; studies of thrombocytopenia in head trauma; and professional society guidelines based on expert opinion. Case Reviews
SC
Two retrospective case reviews were found in the literature review regarding thrombocytopenia and perioperative hemorrhage in neurosurgical patients. Schramm et al., reported on the routine testing of coagulation studies in preoperative neurosurgical patients and only found an
M AN U
association between prolonged APTT and postoperative bleeding.4 No clear association was found for thrombocytopenia, but these authors did not clearly define what threshold was used in order to classify a patient as thrombocytopenic.
Chan et al., conducted a retrospective case review in 1989, which has since been frequently cited, showing a significant association between platelet counts below 100,000/µl with
TE D
postoperative hematoma.5 A failure to respond to transfusion was associated with an increased risk of postoperative hematoma in thrombocytopenic patients. Of note, Chan et al also found that a precipitous drop in platelets in the perioperative period was highly significant: the mean percentage decrease in platelets for patients with hemorrhage was 58.5%, compared with 30.7%
EP
for those without. Greater acute decreases in platelet count, even if the absolute value remained
AC C
above 100,000/µl, were associated with increased risk of postoperative hematoma. Abdelfatah performed a retrospective review of the management of 41 patients with chronic subdural hematomas and intractable thrombocytopenia – defined as platelet count not reaching 100,000/µl after platelet transfusions (18 units of random donor pooled platelets preoperatively, 6 units intaoperatively and 6 units postoperatively).6 Patients had platelet counts ranging from 23,000 to 62,000/µl on admission; the last preoperative platelet count ranged from 51,000 to 79,000/µl. Some had coagulopathies and/or anemia that was also corrected with FFP or packed red blood cell transfusions. The patients in this review were all successfully treated with burr hole evacuations without recurrent SDH. The outcomes of these chronically thrombocytopenia
ACCEPTED MANUSCRIPT
patients suggests that the acuity of a patient's thrombocytopenia may play a role, and that chronically thrombocytopenic patients may be successfully treated with a platelet count below 100,000/µl.
RI PT
NSQIP Three NSQIP database reviews published recently with the goals of evaluating hemostatic
screening and/or thrombocytopenia in neurosurgical patients. Dasenbrock et al., conducted an analysis of NSQIP data with a focus on patients undergoing craniotomy for tumor resection.7
SC
They found that the 30-day mortality rate was significantly higher for patients with moderate (100-124,000/µl), severe (75-99,000/µl) or very severe (<74,000/µl) thrombocytopenia. As well,
M AN U
the rate of unplanned reoperation, especially for postoperative intracranial hemorrhage, was significantly higher for patients with thrombocytopenia classified as severe or very severe. Bronheim et al., conducted an analysis of the NSQIP database with a focus on coagulation parameters in patients undergoing posterior lumbar fusion, one of the most common surgical procedures that routinely requires transfusion.8 Their analyses suggest that a platelet count of <150,000/µl was a significant independent risk factor for transfusion, organ space surgical site
TE D
infection, acute renal failure, reoperation, prolonged ventilation (>48 hours), sepsis, and death. Another NSQIP database analysis, conducted by Seicean et al., focused on the use and utility of preoperative hemostatic screening in adult neurosurgical patients.9 They found a significant
EP
association between abnormal hemostatic parameters and blood transfusions, return to the operating room, and mortality. However, when analyzed individually, platelet count had low
AC C
sensitivity and specificity for predicting any of the aforementioned complications. They did find that an abnormal platelet count was associated with increased intraoperative transfusions, but did not find any significant association with increased bleeding risk. They concluded that preoperative hemostatic screening would be most cost effective when limited to patients with a history of abnormal hemostasis. Trauma Due to the limited evidence found regarding platelet counts and complications in patients undergoing elective neurosurgical procedures, we extended our review of the literature to include
ACCEPTED MANUSCRIPT
investigations on coagulopathies in patients with traumatic injuries involving the central nervous system. Coagulopathy is a well-established predictor of mortality in patients with traumatic brain injury, but investigation on the specific impact of platelets remains scarce.
RI PT
The most notable article in this category is a Cochrane Review conducted by Yuan et al., published in the Journal of Neurotrauma.10 They performed meta-analyses of 19 studies, with a focus on the effects of coagulopathy in patients with traumatic brain injury (TBI), and its effect on progressive hemorrhagic injury (PHI). A small number of studies were included in two meta-
SC
analyses studying the association of thrombocytopenia with the development of TBI; neither was significant. Two published studies failed to show a difference in platelet count when comparing TBI and non-TBI patients. Three studies showed no difference in platelet count in a comparison
M AN U
of isolated TBI patients versus TBI patients with other traumatic injuries. They reviewed fifteen studies comparing coagulopathy in patients with PHI versus those without PHI. They found that patients with evidence of PHI did have a lower platelet count, usually <100,000/µl, and an INR of >1.2. In addition, the authors of this review suggested that the use of thromboelastography for clot strength analysis may be used to more effectively guide transfusion practice, rather than
TE D
simply following quantitative coagulation parameters.
Schnuriger et al., performed a focused retrospective chart review of 310 blunt head trauma patients in order to investigate the impact of platelets on the progression of hemorrhagic injury.11 They found that patients with a platelet count of <100,000/µl had a ninefold increased adjusted
EP
risk of mortality, independent of other coagulation parameters (INR and aPTT). In addition, a platelet count of <175,000/µl was found to be a significant predictor of progression of
AC C
hemorrhagic injury. However, in spite of the association of thrombocytopenia with increased mortality, the platelet count did have a poor sensitivity (11%) for predicting worsening intracranial hemorrhage. Therefore, the applicability of this finding remains uncertain. Bullock et al., reviewed the craniotomy data for 850 patients that underwent evacuation of a post-traumatic intracranial hematoma and identified 59 patients who required a second operation for removal of a second intracranial lesion.12 Coagulopathy was identified as a risk factor associated with post-craniotomy hematoma. More specifically, thrombocytopenia was detected in 19 patients (32%), with platelet counts ranging from 26,000 to 129,000/µl. Serial platelet
ACCEPTED MANUSCRIPT
counts were noted to be lowest on the day after the first craniotomy and in three patients, platelet transfusions were given before their second operation. Thrombocytopenia was not found to be associated with a significantly worse outcome in this case review.
RI PT
Multiple smaller sample size case reviews were published that focused on coagulopathy and the outcome of patients with severe head injury. However, amidst the multiple parameters evaluated for coagulopathy, only limited focus has been placed on platelet count. Gupta et al., did a
prospective case review of 50 patients and did find that thrombocytopenia – defined as platelet count of less than 100,000 – as an independent risk factor for worse clinical outcome in patients
SC
with moderate head injury, but not those with severe head injury.13 Kuo et al., evaluated a group of 61 head injury patients by a coagulopathy score and provided evidence this was a better
M AN U
predictor of mortality than radiographic findings of brain compression.14 However, the mean platelet counts were within the normal range, and only two of 61 patients were thrombocytopenic. Salehpour et al., found that the median serum platelet count was significantly lower in patients with severe isolated head traumas (GCS < 8) who did not survive, although it was >200,000/µl in both groups.15 Konig et al., studied coagulopathy in patients with chronic subdural hematomas and did find a significant difference in patients with recurrent (mean
AC C
Guidelines
EP
subdural hematomas.16
TE D
platelets count 177,000/µl) and non-recurrent (mean platelet count of 226,000/µl) chronic
The American Society of Hematology last published guidelines on platelet transfusion in 2007, the Journal of Hospital Medicine published periprocedural transfusion guidelines in 2013, and the American Association of Blood Banks (AABB) cited various recommendations regarding platelet transfusion thresholds in 2015.1-3 The consensus is that there is poor or low quality data to suggest a platelet transfusion threshold for neuraxial procedures or patients with intracranial hemorrhage. However, due to the potential for significant adverse outcomes, it is expert opinion that platelets should be maintained above 100,000/µl in patients with intracerebral bleeding. More specifically, the AABB recommends prophylactic platelet transfusion for patients
ACCEPTED MANUSCRIPT
undergoing elective diagnostic lumbar punctures with plalelet counts less than 50,000/µl.1 The AABB cannot recommend for or against platelet transfusion for patients with antiplatelet therapy who had intracranial hemorrhage. For patients undergoing nonneuraxial surgery or with nonneuraxial bleeding prophylactic platelet transfusions are recommended if the platelet count is
RI PT
less than 50,000/µl. Discussion
A platelet transfusion threshold of 100,000/µl in patients requiring elective neurosurgical
SC
procedures and in patients with acute neuraxial hemorrhages is a widely accepted standard of practice. However, a review of the literature reveals a paucity of evidence surrounding this topic.
M AN U
The majority of the data consist of retrospective cohort studies or case reviews. In addition, official published guidelines either cite expert opinion or state that they are unable to make formal recommendations regarding the safety of neurosurgical procedures in the setting of thrombocytopenia.1-3
It is understandable that there is a lack of randomized control trials, or even prospective studies, surrounding a platelet transfusion thresholds in patients undergoing elective neurosurgical
TE D
procedures. The risks of intra-operative and post-operative hemorrhage in the confines of the central nervous system carry a much too high morbidity and mortality to justify any such investigation. Thus, this literature review has also explored the trauma literature in an effort to
EP
bolster any evidence for a platelet threshold of 100,000/µl. Some findings in the trauma literature do indicate that thrombocytopenia, with a platelet count of less than 100,000/µl, was associated with increased risk of PHI in traumatic brain injury patients.10 Thrombocytopenia less than
AC C
100,000/µl was also found to be associated with a higher mortality rate.11 The evidence on thrombocytopenia as an independent risk factor alone is still limited as the studies on isolated neuro traumas are few and many polytrauma patients are also at risk for other coagulopathies. 1215
Most articles evaluating the risks of hemorrhage or progressive hemorrhage in the neurosurgical patient cite the retrospective case series conducted by Chan, et al in 1989. The conclusions drawn from this investigation seem to have resonated the most with expert opinion and current practice. Although a platelet transfusion threshold of less than 100,000/µl may be observed as
ACCEPTED MANUSCRIPT
general practice, there are other factors that should be taken into account – platelet transfusion responsiveness; the chronicity of a patient's thrombocytopenia; the acuity and severity of the drop in platelet count as well as the urgency of neurosurgical intervention. In addition, the literature surrounding evaluating patients with coagulopathies is trending towards alternative
RI PT
assessments of coagulability rather than relying purely on absolute platelet count. Elective
neurosurgical oncology patients often suffer from chronic thrombocytopenia due to bone marrow suppression or other etiologies. In these settings, tools, such as thromboelastography, may be
SC
useful in patients with chronic thrombocytopenia. Conclusion
M AN U
It has been standard practice to maintain a patient's platelet count above 100,000/µl in elective neurosurgical patients during the perioperative period. Guidelines regarding platelet transfusion thresholds in the neurosurgical perioperative period are founded on weaker clinical evidence and expert opinion. However, the recommendation of a platelet count greater than 100,000/µl prior to elective neurosurgical procedures is unlikely to be further examined due to the mortality and morbidity associated with increased risk of hemorrhage in the neurosurgical patient. Further
TE D
investigation may be done as to risks of neurosurgical intervention in patients with chronic thrombocytopenia and or those unresponsive to platelet transfusion as well as other tools that
AC C
EP
may be used to assess coagulopathy and bleeding risk in these patients.
References 1. Kaufman RM, Djulbegovic B, Gernsheimer T, Kleinman S, Tinmouth AT, Capocelli KE, Cipolle MD, Cohn CS, Fung MK, Grossman BJ, Mintz PD, O'Malley BA, Sesok-Pizzini DA, Shander A, Stack GE, Webert KE, Weinstein R, Welch BG, Whitman GJ, Wong
ACCEPTED MANUSCRIPT
EC, Tobian AA. Platelet Transfusion: a clinical practice guideline from the AABB. Ann Intern Med. 2015;162:205-213. 2. Slichter, SJ. Evidence-based platelet transfusion guidelines. Hematology Am Soc Hematol Educ Program. 2007:172-178.
blood products. J Hosp Med. 2013 Nov;8(11):647-52.
RI PT
3. Hogshire LC, Patel MS, Rivera E, Carson JL. Evidence review: periprocedural use of
4. Schramm B, Leslie K, Myles PS, Hogan CJ. Coagulation studies in preoperative neurosurgical patients. Anaesth Intensive Care 2001; 29:388-392.
SC
5. Chan KH, Mann KS, Chan TK. The significance of thrombocytopenia in the development of postop intracranial hematoma. J Neurosurg 71:38-41, 1989.
M AN U
6. Abdelfatah, M. Management of chronic subdural hematoma in patients with intractable
thrombocytopenia. Turk Neurosurg; 2016.
7. Dasenbrock HH, Devine CA, Liu KX, Gormley WB, Claus EB, Smith TR, Dunn IF. Thrombocytopenia and craniotomy for tumor: a NSQIP analysis. Cancer 2016;122:170817.
8. Bronheim RS, Oermann EK, Cho SK, Caridi JM. Coagulation profile as a risk factor for
TE D
30-day morbidity and mortality following posterior lumbar fusion. Spine 2016. 9. Seicean A, Schiltz NK, Seicean S, Alan N, Neuhauser D, Weil RJ. Use and utility of preoperative hemostatic screening and patient history in adult neurosurgical patients. J Neurosurg 116:1097-1105, 2012.
EP
10. Yuan Q, Sun YR, Wu X, Yu J, Li ZQ, Du ZY, Wu XH, Zhou LF, Hu J. Coagulopathy in Traumatic Brain Injury and Its Correlation with Progressive Hemorrhagic Injury: A
AC C
Systematic Review and Meta-Analysis. J Neurotrauma. 2016 Jul 15;33(14):1279-91. 11. Schnüriger B1, Inaba K, Abdelsayed GA, Lustenberger T, Eberle BM, Barmparas G, Talving P, Demetriades D. The impact of platelets on the progression of traumatic intracranial hemorrhage. J Trauma. 2010 Apr;68(4):881-5.
12. Bullock R, Hannemann CO, Murray L, Teasdale GM. Recurrent hematomas following craniotomy for traumatic intracranial mass. J Neurosurg 72:9-14, 1990. 13. Gupta G, Wadhwa C, Garg R, Dhaiya RS, Kaushal RK. Impact of Coagulation Profile on Outcome of Head Injury. J Clin Diagn Res. 2016 Jan;10(1):PC04-6. 14. Kuo JR1, Chou TJ, Chio CC. Coagulopathy as a parameter to predict the outcome in head
ACCEPTED MANUSCRIPT
injury patients--analysis of 61 cases. J Clin Neurosci. 2004 Sep;11(7):710-4. 15. Salehpour F1, Bazzazi AM, Porhomayon J, Nader ND. Correlation between coagulopathy and outcome in severe head trauma in neurointensive care and trauma units. J Crit Care. 2011 Aug;26(4):352-6.
RI PT
16. König SA1, Schick U, Döhnert J, Goldammer A, Vitzthum HE. Coagulopathy and outcome in patients with chronic subdural haematoma. Acta Neurol Scand. 2003
AC C
EP
TE D
M AN U
SC
Feb;107(2):110-6.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
NSQIP: National Surgical Quality Improvement Program TBI: traumatic brain injury PHI: progressive hemorrhagic injury INR: international normalized ratio aPTT: activated partial thromboplastin time AABB: American Association of Blood Banks
RI PT
ABBREVIATIONS
ACCEPTED MANUSCRIPT
Disclosure
This research did not receive any specific grant from funding agencies in the public,
RI PT
commercial, or not-for-profit sectors. The authors have no financial conflict of interst to
AC C
EP
TE D
M AN U
SC
disclose.