E Brief Report

Neuraxial Anesthesia in Parturients with Low Platelet

Counts
Jeffrey Bernstein, MD,* Betty Hua, MD,† Madelyn Kahana, MD,* Naum Shaparin, MD,*
Simon Yu, MD,* and Juan Davila-Velazquez, MD*

The obstetric anesthesiologist must consider the risk of spinal–epidural hematoma in patients
with thrombocytopenia when choosing to provide neuraxial anesthesia. There are little data
exploring this complication in the parturient. In this single-center retrospective study of 20,244
obstetric patients, the incidence of peripartum thrombocytopenia (platelet count <100,000/mm3)
was 1.8% (368 patients). Of these patients, 69% (256) received neuraxial anesthesia.
No neuraxial hematoma occurred in any of our patients. The upper 95% confidence limit for
spinal–epidural hematoma in patients who received neuraxial anesthesia with a platelet count
of <100,000/mm3 was 1.2%. (Anesth Analg 2016;123:165–7)

N
euraxial anesthesia is the most effective modal- upper 95% confidence limit of the incidence of neuraxial
ity for labor analgesia and the standard of care for hematoma after neuraxial anesthesia using data from our
most cesarean deliveries. For the parturient with institution and after combining our data with previously
thrombocytopenia, the risk of general anesthesia must published data.
be balanced with the risk of neuraxial hematoma, a risk
enhanced by decreased platelet number, function, or both. METHODS
Mild thrombocytopenia as defined by a platelet count After approval from our IRB, the need for written informed
of between 100,000 and 150,000/mm3 occurs with an esti- patient consent was waived. The patented hospital software
mated peripartum frequency of 6.4%.1 Thrombocytopenia, program, Clinical Looking Glass, was used to identify par-
defined as a platelet count of <100,000/mm3 platelets, is turients who delivered at our hospital between September
estimated to occur in 1% of parturients.2 Primary hemo- 1, 2009, and September 1, 2013, with a documented platelet
static function can be compromised by decreased numbers count of <100,000/mm3 at admission and up until 5 days
of platelets with preserved function, as observed in patients after delivery. At our institution, we obtained a complete
with idiopathic thrombocytopenic purpura or gestational blood count, including a platelet count, on admission to
thrombocytopenia,3 and by impaired platelet function as the labor and delivery suite. Two investigators performed a
observed in some patients with thrombocytopenia associ- detailed review of the medical record of all included study
ated with hypertensive disorders in pregnancy. Studies subjects. We recorded the etiology of the thrombocytopenia,
evaluating platelet function in pre-eclamptic parturients the type of anesthetic technique (epidural, spinal, general,
using thromboelastography and a platelet function analyzer or no anesthesia), the mode of delivery (vaginal or cesarean
have shown preservation of hemostatic function with plate- delivery), and any major neurologic and anesthetic compli-
let counts as low as 54,000/mm3.3–5 It has been suggested by cations during hospitalization. We also determined whether
a number of authors4–7 that neuraxial anesthesia is safe in the platelet counts of ≤100,000/mm3 were known before initia-
parturient if the platelet count is higher than 75,000/mm3. tion of neuraxial anesthesia, as documented on the anes-
Tanaka et al.3 suggested a lower limit of 50,000/mm3 thesia preoperative evaluation and confirmed by timed
platelets before neuraxial anesthesia is considered unsafe, laboratory results in the medical record.
excluding pre-eclamptic patients. The upper limit of the 95% confidence interval (CI) for
The purpose of our retrospective study was to determine the incidence of neuraxial hematoma was calculated using
the incidence and etiology of thrombocytopenia in a large the Rule of 3 and the formula R = 1 − [(0.05)^(1/N)]. R is an
data set of parturients at a single center, describe the inci- estimate of the upper limit of the 95% CI for the risk of peri-
dence of spinal–epidural hematoma after neuraxial anes- partum epidural hematoma and N is the number of subjects
thesia in the thrombocytopenic patient, and estimate the without such an event in the cohort. We estimated the upper
bound of risk twice, first using the number of patients from
our cohort and second by pooling the patients together
From the *Department of Anesthesiology, Montefiore Medical Center, from previously published studies.
Albert Einstein College of Medicine, Bronx, New York; and †Department
of Anesthesiology, Stony Brook University Medical Center, Stony Brook,
New York. RESULTS
Accepted for publication February 29, 2016. During the study period, 20,244 parturients were identified
Funding: None. from the database, of whom 368 (1.8%; 95% CI, 1.6%–2.0%)
The authors declare no conflicts of interest. had platelet counts of ≤100,000/mm3 during the peripar-
Reprints will not be available from the authors. tum period. The most common cause of thrombocytopenia,
Address correspondence to Juan Davila-Velazquez, MD, Montefiore Medical before or after delivery, was gestational thrombocytopenia
Center/AECOM, 1825 Eastchester Rd., Bronx, NY 10461. Address e-mail to
judavila@montefiore.org.
(Table 1). In 256 parturients, platelet counts were known to
Copyright © 2016 International Anesthesia Research Society be low before anesthetic management decisions, including
DOI: 10.1213/ANE.0000000000001312 placement of a neuraxial anesthetic. In 112 patients, the low

July 2016 • Volume 123 • Number 1 www.anesthesia-analgesia.org 165

Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
 E Brief Report

Table 1.  Etiology of Thrombocytopenia

Platelet count Platelet count
Etiology <80,000/mm3 (n) >80,000/mm3 (n) Total n (%)
Pre-eclampsia 32 57 89 (24)
Gestational thrombocytopenia 37 138 175 (48)
Idiopathic thrombocytopenic purpura 22 15 37 (10)
Othera 11 56 67 (18)
n 102 266 368
Unknown or unclear reasons.
a

platelet count was identified only after neuraxial anesthesia of coagulopathy, abnormal bruising, or bleeding. There are
placement and within 24 hours after delivery. 4 previous major studies in parturients that categorized a
Of the 256 parturients with low platelet counts identi- subset of patients with a platelet count of <100,000/mm3.
fied before delivery, 151 (59%) received neuraxial anesthesia; The studies by Beilin et al.,9 Frenk et al.,11 Tanaka et al.,3
94% (105/112) of the patients in whom thrombocytopenia and Goodier et al.7 included 80, 177, 75, and 173 patients,
was unknown at the time of placement, but diagnosed after respectively. Overall, there were no reports of neurologic
delivery, received neuraxial anesthesia. Approximately 60% sequelae secondary to the initiation of neuraxial anesthe-
of thrombocytopenic parturients who received neuraxial sia in any of these data sets. The incidence of platelet count
anesthesia had an epidural or a combined spinal–epidural <100,000/mm3 was 80 of 15,919 (0.50%; 95% CI, 0.30%–60%)
technique. Forty-four percent of the patients with thrombo- in the study by Beilin et al.9 and 252 of 52,000 (0.54%; 95%
cytopenia of <100,000/mm3 underwent cesarean delivery (at CI, 0.47%–59%) in the study by Goodier et al.7 The overall
our institution, the overall cesarean delivery rate is approxi- incidence of thrombocytopenia was not reported in study
mately 30%). There were no neurologic complications in any by Frenk et al.11 The study by Tanaka et al.3 excluded par-
of the patients who received a neuraxial anesthetic. turients with pre-eclampsia; hence, the true incidence is
We calculated the 95% risk estimate by combining our not known. Our incidence of thrombocytopenia (1.8%)
results with the 4 other studies describing neuraxial anes- was almost 4 times more than the incidence found in pre-
thetics in thrombocytopenic parturients. Before the recent vious studies; this may be because of a greater number of
publication of the study by Goodier et al.7, 326 total cases of high-risk parturients with pre-eclampsia. In the studies by
neuraxial anesthesia were reported in small-to-moderately Beilin et al.,9 Frenk et al.,11 Tanaka et al.,3 and Goodier et al.,7
sized studies.3,8,9 Combining the patients from Goodier neuraxial anesthesia was performed in 65%, 96%, 63%, and
et al.7 (173) and the patients in the current study (256) to 63%, respectively, of thrombocytopenic patients. We per-
these patients resulted in a final composite sample size of formed neuraxial blocks on 68% (256/368) of patients with
755. No patient experienced a spinal–epidural hematoma, a thrombocytopenia, a rate similar to that described by other
zero numerator. The upper limit of the 95% CI for the risk of investigators.3,7,9
spinal–epidural hematoma from our study of 256 patients Beilin et al.9 and Goodier et al.7 recommended evaluation
was 0.012 or 1.2% and of all cohorts together was 0.004 or of the risk–benefit ratio for each individual before initiation
0.4%. Given the very few patients with platelet counts of of neuraxial anesthesia. No previously published study has
<50,000 mm3, meaningful statistical analysis for this group had a large enough population to determine the safety of
is not possible. neuraxial anesthesia placement in parturients with platelet
counts <100,000/mm3.
DISCUSSION Similar to the practice patterns described by Beilin
The major concern of the anesthesiologist in placing neurax- et al.,12 most of our anesthesiologists felt comfortable plac-
ial anesthesia in a patient with thrombocytopenia is an epi- ing neuraxial anesthesia in otherwise healthy parturi-
dural or spinal hematoma. The risk of epidural hematoma ents with stable platelet counts between 80,000/mm3 and
must be weighed against the risk of general anesthesia. The 100,000/mm3, typically observed in patients with gesta-
only complication observed at our institution attributable tional thrombocytopenia. However, additional concern is
to general anesthesia in the obstetric population during this present in patients with pre-eclampsia in whom both the
time period was a case of dental injury in a patient who did number of platelets may decrease precipitously and the
not have thrombocytopenia. No mortality in thrombocyto- quality of the platelets may also be impaired.13–15 Ideally, in
penic patients receiving general anesthesia was reported. these patients, a test of platelet function such as thrombo-
The development of a neuraxial hematoma after the place- elastography is performed before central neuraxial block
ment of a neuraxial block is rare and has been estimated at placement.
1:150,000 after epidural and 1:220,000 after placement of a Our study was limited by its retrospective nature. The
spinal anesthetic in the general population.10 In the obstet- anesthetic management was dictated by the individual
ric population, the incidence of neuraxial hematoma is attending anesthesiologists. Tests of platelet function were
estimated at 1:168,000 after the administration of epidural not consistently performed. Additional large studies in
analgesia.9 The incidence of hematoma after spinal anesthe- pregnant patients with thrombocytopenia and with platelet
sia in the obstetric patient is unknown. dysfunction who receive neuraxial anesthesia will help to
To the best of our knowledge, there are no confirmed build on the existing database to further support safe rec-
case reports of hematomas after neuraxial anesthesia in ommendations and limitations of neuraxial techniques in
parturients with thrombocytopenia, absent clinical signs the parturient with low platelet counts. E

166   
www.anesthesia-analgesia.org anesthesia & analgesia
Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
 Regional Anesthesia Low Platelet Counts

DISCLOSURES 3. Tanaka M, Balki M, McLeod A, Carvalho JC. Regional anesthe-

Name: Jeffrey Bernstein, MD. sia and non-preeclamptic thrombocytopenia: time to re-think
Contribution: This author helped design the study, conduct the the safe platelet count. Rev Bras Anestesiol 2009;59:142–53
study, analyze the data, and write the manuscript. 4 Orlikowski CE, Rocke DA, Murray WB, Gouws E, Moodley
J, Kenoyer DG, Byrne S. Thrombelastography changes in pre-
Attestation: Jeffrey Bernstein has seen the original study data, eclampsia and eclampsia. Br J Anaesth 1996;77:157–61
reviewed the analysis of the data, approved the final manuscript, 5. Vincelot A, Nathan N, Collet D, Mehaddi Y, Grandchamp P,
and is the author responsible for archiving the study files. Julia A. Platelet function during pregnancy: an evaluation
Name: Betty Hua, MD. using the PFA-100 analyser. Br J Anaesth 2001;87:890–3
Contribution: This author helped design the study, conduct the 6. Douglas MJ. Platelets, the parturient and regional anesthesia.
study, and analyze the data. Int J Obstet Anesth 2001;10:113–120
7 Goodier CG, Lu JT, Hebbar L, Segal BS, Goetzl L. Neuraxial
Attestation: Betty Hua has seen the original study data,
anesthesia in parturients with thrombocytopenia: a multisite
reviewed the analysis of the data, and approved the final retrospective cohort study. Anesth Analg 2015;121:988–91
manuscript. 8. Ruppen W, Derry S, McQuay H, Moore RA. Incidence of epi-
Name: Madelyn Kahana, MD. dural hematoma, infection, and neurologic injury in obstetric
Contribution: This author helped write the manuscript. patients with epidural analgesia/anesthesia. Anesthesiology
Attestation: Madelyn Kahana approved the final manuscript. 2006;105:394–9
Name: Naum Shaparin, MD. 9. Beilin Y, Zahn J, Comerford M. Safe epidural analgesia in thirty
parturients with platelet counts between 69,000 and 98,000
Contribution: This author helped write the manuscript. mm(-3). Anesth Analg 1997;85:385–8
Attestation: Naum Shaparin approved the final manuscript. 10. Vandermeulen EP, Van Aken H, Vermylen J. Anticoagulants
Name: Simon Yu, MD. and spinal–epidural anesthesia. Anesth Analg 1994;79:1165–77
Contribution: This author helped write the manuscript. 11. Frenk V, Camann W, Shankar KB. Regional anesthesia in partu-
Attestation: Simon Yu approved the final manuscript. rients with low platelet counts. Can J Anaesth 2005;52:114
Name: Juan Davila-Velazquez, MD. 12. Beilin Y, Bodian CA, Haddad EM, Leibowitz AB. Practice pat-
terns of anesthesiologists regarding situations in obstetric
Contribution: This author helped write the manuscript.
anesthesia where clinical management is controversial. Anesth
Attestation: Juan Davila-Velazquez approved the final Analg 1996;83:735–41
manuscript. 13. Sharma SK, Philip J, Whitten CW, Padakandla UB, Landers
This manuscript was handled by: Cynthia A. Wong, MD. DF. Assessment of changes in coagulation in parturients with
preeclampsia using thromboelastography. Anesthesiology
REFERENCES 1999;90:385–90
1. Rolbin SH, Abbott D, Musclow E, Papsin F, Lie LM, Freedman 14. Kelton JG, Hunter DJ, Neame PB. A platelet function defect in
J. Epidural anesthesia in pregnant patients with low platelet preeclampsia. Obstet Gynecol 1985;65:107–9
counts. Obstet Gynecol 1988;71:918–20 15. Davies JR, Fernando R, Hallworth SP. Hemostatic function in
2. Boehlen F, Hohlfeld P, Extermann P, de Moerloose P. Maternal healthy pregnant and preeclamptic women: an assessment
antiplatelet antibodies in predicting risk of neonatal thrombo- using the platelet function analyzer (PFA-100) and thrombo-
cytopenia. Obstet Gynecol 1999;93:169–73 elastograph. Anesth Analg 2007;104:416–20

July 2016 • Volume 123 • Number 1 www.anesthesia-analgesia.org 167

Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.