Neurology

An unexpected effect of traumatic and bloody lumbar puncture among patients with primary headaches in emergency department

a b s t r a c t

Background: This study aimed to investigate the association between traumatic tap and the incidence of post- dural puncture headache (PDPH) following lumbar puncture among patients who underwent LP with a primary discharge diagnosis of primary headache in the emergency department (ED).

Methods: We retrospectively reviewed the medical records of patients who visited a single Tertiary ED with the symptom of a headache and underwent LP for cerebrospinal fluid (CSF) analysis between January 2012 and Jan- uary 2022. Patients who met the definition of PDPH and revisited the ED or outpatient clinic within 2 weeks of discharge were included. For comparative analysis, we divided the groups according to CSF RBC counts (group 1, CSF RBC <10 cells/uL; group 2, 10-100 cells/uL; group 3, >=100 cells/uL). The primary outcome was the differ- ence in CSF RBC counts between the ED or outpatient clinic revisiting patients who underwent LP within 2 weeks after discharge from the ED. The secondary outcomes were the admission rate and risk factors for PDPH; sex, age, needle size, and CSF pressure.

Results: Data from 112 patients were collected; PDPH was reported in 39 patients (34.8%), and 40 (35.7%) patients were admitted. The median (interquartile range) CSF RBC count was 10 [2-100.8] cells/uL. One-way analysis of variance test of the mean differences among the three groups showed no differences in age, the duration of headache before LP, PLT counts, PT, or aPTT among the groups. There were differences in the number of admitted patients (30 vs. 7 vs. 3, P < 0.001) and the incidence of PDPH (29 vs. 6 vs. 4, P < 0.003). In the com- parison of the PDPH and non-PDPH groups, there were differences in age (28.7 +- 8.4 years vs. 36.9 +- 18.4 years, P = 0.01) and the admission rate (85% vs. 9%, P < 0.001).

Conclusions: Notably, our results suggest that traumatic LP may be an unexpected factor in reducing the occur- rence rate of PDPH. Consequently, the admission rate for PDPH was significantly reduced among patients with traumatic LP and those with primary headaches. In this study, we collected and analyzed the data from a relatively small sample size of 112 patients. Further studies are needed to evaluate the relationship between trau- matic LP and PDPH.

(C) 2023 Published by Elsevier Inc.

List of abbreviations

PLT platelet

PT prothrombin time

aPTT activated partial thromboplastin time

ED emergency department

LP lumbar puncture

CSF cerebrospinal fluid

PDPH post-dural puncture headache

CSF cerebrospinal fluid

EBPs epidural blood patches

CT computed tomography

G gauge

RBC red blood cells

* Corresponding authors.

E-mail addresses: [email protected] (Y. Ko), [email protected] (H. Yang).

1 Heewon Yang and Yura Ko contributed equally to this work.

ABR absolute bed rest

IQR interquartile range

vs. versus

ANOVA analysis of variance

  1. Introduction

To diagnose the cause of headaches in the emergency department (ED), lumbar puncture is crucial to analyze cerebrospinal fluid (CSF) to differentiate causes of secondary headaches, including central nervous system infection, autoimmune diseases, inflammation, and Brain hemorrhage [1]. Although LP requires simple manipulation skills,

https://doi.org/10.1016/j.ajem.2023.03.038 0735-6757/(C) 2023 Published by Elsevier Inc.

the procedure results in the tearing of the spinal meningeal layer (dura mater) and injury of the epidural space, which consists of loose connec- tive tissue, adipose tissue, and internal venous plexus [2].

LP results in limited complications, including puncture site infection, bleeding, and post-dural puncture headache (PDPH) [1]. PDPH is de- fined as any headache after an LP that worsens within 15 min of sitting or standing and is relieved within 15 min of lying down [1]. Although the pathophysiology of PDPH is still not fully understood, it is well known that puncture of the spinal meningeal layer allows cerebro- spinal fluid (CSF) to leak from the subarachnoid space, resulting in a decrease in CSF volume and pressure [3]. This CSF volume loss may cause a downward pull on pain-sensitive structures, resulting in postural headaches [4].

Traumatic LP refers to the presence of blood in the CSF and occurs in 10-20% of cases [5,6]. In 1960, Gormley et al. attempted a novel epidural blood patch (EBP) to treat PDPH; they had previously thought that bloody taps might be associated with a lower incidence of PDPH [7,8]. Considering the hypothesis that the occurrence of PDPH would be reduced by preventing CSF leakage by coagulation with a mechanism similar to that of an EBP in patients with traumatic LP, we investigated the relationship between PDPH and traumatic LP [9].

The primary outcome was the difference in CSF RBC counts between the ED or outpatient clinic revisiting patients who underwent LP within 2 weeks after discharge from the ED. The secondary outcomes were the admission rate and risk factors for PDPH; needle size, sex, age, diabetes mellitus, hypertension, Anticoagulation medication, platelet count, prothrombin time , Activated partial thromboplastin time , and CSF pressure.

  1. Methods
    1. Study design and population

This study retrospectively collected data from patients who visited a single tertiary ED with the symptom of a headache and had undergone diagnostic LP for CSF analysis. Standardized extraction of demographic, clinical, and laboratory data from medical records was performed, and the data were reviewed and analyzed between January 2012 and Janu- ary 2022. The study was approved by the institutional review board of the Ajou university medical center (AJOUIRB-DB-2022-511).

To evaluate PDPH after diagnostic LP, we followed up with the pa- tients who met the definition of PDPH on their revisits to the ED or out- patient clinic among the patients discharged not those who experience PDPH immediately after LP. The exclusion criteria were age < 15 years, secondary headache including meningoencephalitis (infection, inflam- mation, and autoimmune diseases), hemorrhage, vessel malformation, increased intracranial pressure, and head and Neck trauma. Primary headache was defined as a headache that was not caused by a structural problem of the brain and secondary headache as a headache caused by a structural problem in the brain [10].

One-way ANOVA and posthoc analysis “>In our ED, emergency physicians with at least 1 year of diagnostic LP experience performed the diagnostic LP procedure. Before performing LP, all patients underwent brain computed tomography (CT) to rule out brain hemorrhage or increased intracranial pressure. When per- forming LP, the patient was maintained in the lateral decubitus position. The cutting Quincke needles (atraumatic needles were not available in our hospital) and 20-23 gauge (G) needles were used. The needle bevel axis was kept parallel to the spinal axis, and the procedure was performed at the L3/4 level. We used the gravity drip method to collect the CSF and reinserted the style of the spinal puncture needle before re- moval to prevent additional injury to the meninges. Lying supine with- out a pillow cushion for 6-8 h following LP remains a patient requirement of standard operating procedures in our hospital.

We regularly collected samples for CSF analysis, and the opening pressure was measured before collecting the first tube. In our ED, spinal tap notes were recorded, including the needle size, initial opening pressure, bedrest time, and other complications.

    1. Traumatic lumbar puncture and outcome measurement

Traumatic LP is used to evaluate the CSF’s red blood cells (RBC). To differentiate subarachnoid hemorrhage , 2000 x 106/uL CSF RBC was used as the cut-off value based on previous studies [11]. The 3- tube method was used to discriminate SAH for those with more RBC in the CSF. Because the RBC count rapidly decreased to 0 in the third tube, we analyzed the first CSF tube.

As CSF RBC counts have a wide range from 0 to 10,000 cells/uL, we

converted these values using the log function equation for approximat- ing them a normal distribution: log10 (CSF RBC + 1). For comparisons among the group, we divided them according to CSF RBC counts as fol- lows: group 1, CSF RBC <10 cells/uL; group 2, 10-100 cells/uL; group 3,

>=100 cells/uL.

    1. Statistical analysis and sample size estimation

Continuous data are expressed as mean +- standard deviation or me- dian (interquartile range [IQR]) based on the results of the normal dis- tribution analysis. Where appropriate, categorical data are presented as absolute values, together with frequency distributions. To convert the log formula according to the CSF RBC count, we divided the patients into three subgroups. After the normality test, one-way analysis of var- iance (ANOVA) and post hoc analysis with Bonferroni correction were used to compare variables between the subgroups and groups, respec- tively. The Chi-squared test or Fisher’s exact test was used to analyze categorical variables, as appropriate. Data analysis was conducted using the Statistical Package For The Social Sciences 25 (IBM Corp., Armonk, NY, USA). Statistical significance was set at p < 0.05.

According to the randomized controlled trial in 2020 and the pro-

spective observational cohort study in 2022, 32.8% and 29.5% of patients developed PDPH after LP [3,16], and accordingly, the adequate mini- mum sample size for a sufficient power was 64 and 44 patients, respec- tively (80% power; ? =0.05).

  1. Results
    1. General characteristics of the patientswb

Data were collected from 112 patients (Table 1): male, 54 (48.2%);

mean age, 34 +- 16.1 years. PDPH was reported in 39 patients (34.8%), and 40 patients (35.7%) were admitted. The median [IQR] CSF RBC count was 10 [2-101] cells/uL, and the median [IQR] CSF white blood cell count was 1.0 [0-2] cells/uL. The mean CSF protein level was

37.8 +- 25.9 mg/mL, and the mean CSF glucose level was 62.7 +-

12.8 mmol/L. The mean opening pressure was 16.1 +- 5.1 mmHg, and mean log (CSF RBC + 1) was 1.3 +- 1.1.

    1. One-way ANOVA and posthoc analysis of data between the three groups according to the CSF RBC counts

Table 2 presents the results of the one-way ANOVA test of the mean differences between the three groups according to the CSF RBC counts. Group 1 had RBC <10 cells/uL, group 2 had 10 <= RBC < 100 cells/uL, and group 3 had CSF RBC >=100 cells/uL. The numbers of patients in each group were 59, 25, and 28, respectively. There were no differences in age, the duration of headache before LP, PLT counts, PT, or aPTT among the groups. There was a significant association between the CSF analysis results including Log10 (CSF RBC + 1) (0.4 +- 0.3 versus [vs.] 1.6 +- 0.3 vs. 3.0 +- 0.8, P < 0.001) and CSF protein levels (32.5 +- 13.5 vs. 35.4 +- 16.0 vs. 50.9 +- 43.4, P < 0.006) among the three groups.

Table 1

Baseline characteristics.

Patient’s characteristics (n = 112)

Table 3

A comparison of cerebrospinal fluid Red blood cell counts and post-dural puncture head- ache after diagnostic lumbar puncture.

Age (years) 34.0 +- 16.1

Female (n,%) 58 (51.8)

Group 1

(RBC < 10)

Group 2

(10 <= RBC < 100)

Group 3

(RBC >= 100)

P-value

Diabetes mellitus (n,%) 5 (4.5)

Hypertension (n,%) 6 (5.4)

Stroke (n,%) 3 (2.7)

anticoagulation drug (n,%) 1 (0.9)

Laboratory test

Hemoglobin (g/dL) 13.5 +- 2.0

White blood cell (*1000/uL) 9.1 +- 4.3

Platelet (*1000/uL) 220.9 +- 73.6

Prothrombin time (INR) 1.1 +- 0.2

Activated partial thromboplastin time (second) 32.8 +- 7.7

Initial lumbar puncture pressure 16.1 +- 5.1 Cerebrospinal fluid analysis

Red blood cell (cells/uL) 10.0 [2-101]

White blood cell (cells/uL) 1.0 [0-2]

Protein (mg/mL) 37.8 +- 25.9

Glucose (mmol/L) 62.7 +- 12.8

Log (CSF RBC + 1) 1.3 +- 1.1

Post-dural puncture headache (n,%) 39 (34.8)

Admission (n,%) 40 (35.7)

Data is presented as mean +- standard deviation or median [Interquartile range] for the continuous variables and number (%) for the categorical variables. DM; diabetes mellitus, HTN; hypertension, CAOD; coronary artery obstructive disease, PDPH; Postdural puncture headache.

Bonferroni’s posthoc analysis showed only differences in log (CSF RBC + 1) among the groups.

    1. Comparison of characteristics between the subgroups

Table 3 presents a comparison of CSF RBC counts and PDPH after di- agnostic LP. There were no differences in sex, underlying chronic dis- eases, or needle sizes among the groups. However, there were differences in the number of admitted patients (30 vs. 7 vs. 3, P < 0.001) and the incidence of PDPH (29 vs. 6 vs. 4, P < 0.003) among the groups.

Table 4 presents the comparison of data between the PDPH and non- PDPH groups (39 vs. 73 patients). There were age differences (28.7 +- 8.4 vs. 36.9 +- 18.4, P = 0.01), and the admission rate (85% vs. 9%, P < 0.001) between the two groups. PDPH occurrence rate according to CSF RBC counts (n, %) was RBC < 10 (29, 49.2%), 10 <= RBC < 100 (6,

24%), 100 <= RBC < 1000 (4, 25%), and RBC >= 1000 (0%) (P = 0.002).

  1. Discussion

Previous studies have documented that the incidence of PDPH is

3.5-33% after LP [12]. Several studies have shown risk factors for PDPH. Common factors influencing the incidence of PDPH include

Sex 0.35

P

Male

25(46.3)

15(27.8)

14(25.9)

Female

34(58.6)

10(17.2)

14(24.1)

ast Hx

DM 1 1 2 0.53

HTN

3

1

2

1.00

Stroke

1

0

2

1.00

Admission

30(75)

7(17.5)

3(7.5)

<0.001

PDPH

29(74.4)

6(15.4)

4(10.3)

<0.001

Needle size? 0.73

20G

2(66.7)

0(0)

1(33.3)

21G

12(52.2)

3(13.0)

8(34.8)

22G

43(51.2)

22(26.2)

19(22.6)

23G

1(100)

0(0)

0(0)

Data are presented as number (%) for the categorical variables. DM; diabetes mellitus, HTN; hypertension, CAOD; coronary artery obstructive disease, PDPH; post-dural punc- ture headache.

* The Chi-squared test and Fisher’s exact test for needle size and cerebrospinal fluid

RBCs counts (n = 111)

female sex, age 20-50 years, and history of PDPH or headache [3,13]. The risk factors for the procedure include the needle size, design, and orientation and style of needle reinsertion [14]. To prevent PDPH, a non-traumatic and non-amputation needle has been suggested as the most effective intervention [12,15]. While researchers have docu- mented methods for preventing PDPH, their studies focused on the risk factors for PDPH, regardless of primary and secondary headaches. Our study aimed to investigate the association between traumatic bloody tap and the incidence of PDPH following LP among patients who underwent LP in the ED for over 10 years.

To exclude brain hemorrhage, the patients underwent brain CT be- fore diagnostic LP. No secondary headache events were initially misdiagnosed in the ED. Based on clinical symptoms, 39 patients were diagnosed with PDPH. Previous studies support our results that patients with PDPH were younger than those without PDPH. A previous study indicated that needle sizes of 21- and 25-G Sprotte needles were related to the occurrence rate of PDPH [16]. As we usually used 21- and 22-G Quinkle-type needles, our results showed no difference in the occur- rence of traumatic LP and PDPH with the needle size.

There was no PDPH occurance in patients with CSF RBC counts

>=1000 (Table 4), suggesting that the greater the CSF RBC count, the lower the occurrence of PDPH after LP. Furthermore, patients whose CSF RBC count was <10 showed a remarkable occurrence rate of PDPH. In this regard, our findings suggested that bloody taps may be associated with a lower incidence of PDPH.

Table 2

One-way ANOVA test of mean difference according to cerebrospinal fluid red blood cell counts.

Groups

P value

Posthoc P-value

1 (RBC < 10)

2 (10 <= RBC < 100)

3 (RBC >= 100)

1 vs 2

1 vs 3

2 vs 3

Number (n =112)

59

25

28

Age

34.4 +- 16.3

30.8 +- 12.5

36.1 +- 18.5

0.47

1.00

1.00

0.70

Onset to tap (hr)

54.5 +- 62.6

61.5 +- 57.7

50.3 +- 49.2

0.78

1.00

1.00

1.00

CBC_WBC (1000/uL)

9.9 +- 4.7

8.8 +- 3.7

8.0 +- 3.3

0.15

0.90

0.17

1.00

CBC_Hb (g/dL)

13.4 +- 1.5

14.2 +- 1.3

13.1 +- 3.0

0.11

0.25

1.00

0.14

CBC_PLT (1000/uL)

217.9 +- 69.4

232.6 +- 82.2

216.5 +- 75.7

0.69

1.00

1.00

1.00

INR

1.1 +- 0.1

1.0 +- 0.3

1.0 +- 0.3

0.74

0.48

0.09

1.00

aPTT (second)

33.9 +- 5.3

31.4 +- 11.1

31.9 +- 8.2

0.38

0.65

0.94

1.00

CSF_PTN (mg/mL)

32.5 +- 13.5

35.4 +- 16.0

50.9 +- 43.4

0.006

1.00

0.005

0.08

CSF_GLU (mmol/L)

62.3 +- 9.2

65.6 +- 21.4

60.8 +- 8.7

0.37

0.84

1.00

0.51

Log (RBC + 1)

0.4 +- 0.3

1.6 +- 0.3

3.0 +- 0.8

<0.001

<0.001

<0.001

<0.001

Data is presented as mean +- standard deviation. Statistical comparations (ANOVA) of measurements among different groups, and inter-group comparison (Bonferroni posthoc test), sig- nificant, P < 0.05. CBC; complete blood count, WBC; white blood cell, PLT; platelet, aPTT; activated partial thromboplastin time, PTN; protein, GLU;glucose.

Table 4

A comparison of post-dural puncture headache (PDPH) and non-PDPH groups.

PDPH

no-PDPH

P value

Number

39

73

Sex

0.938

Male

19(35.2)

35(64.8)

Female

20(34.5)

38(65.5)

Age

28.7 +- 8.4

36.9 +- 18.4

0.010

Height

169.3 +- 8.5

164.6 +- 10.8

0.279

Weight

Cerebrospinal fluid

64.9 +- 12.3

68.1 +- 14.9

0.528

0.002

RBC < 10

29(49.2)

30(50.8)

10 <= RBC < 100

6(24.0)

19(76.0)

100 <= RBC < 1000

4(25.0)

12(75.0)

RBC >= 1000

0(0)

12(100)

LP Pressure

16.98 +- 3.9

15.7 +- 5.6

0.167

Admission

33

7

< 0.001

Needle size?

0.053

20G

0

3(100)

21G

4(17.4)

19(82.6)

22G

34(40.5)

50(59.5)

23G

1(100)

0(0)

Data are presented as mean (standard deviation) for the continuous variables and number (%) for the categorical variables. LP; lumbar puncture, RBC; red blood cell.

* The Chi-squared test and Fisher’s exact test for needle size and PDPH occurance

(n = 111)

This study has several limitations. First, the diagnosis of PDPH was made using only clinical symptoms without other imaging modalities, including magnetic resonance imaging to detect CSF leakage [17]. How- ever, we only included patients with primary headaches, and PDPH can be diagnosed based on clinical symptoms with characteristics of postural-dependent patterns [18]. Second, Patients lost to follow-up were not included in our study. However, in our ED, After the diagnostic LP, all patients received patient education, including the possibility, symptoms, and prevention methods of PDPH. Moreover, all patients were scheduled for Short-term outpatient follow-ups. If the patient did not visit the outpatient clinic, the medical staff forwarded the pa- tient a follow-up telephone survey. Third, our findings cannot be ap- plied to all patients because of the presence of secondary headaches, including those due to increased intracranial pressure or structural brain lesions. However, to the best of our knowledge, this is a novel in- vestigation of traumatic LP in patients with primary headaches. Finally, we collected and analyzed the data from a relatively small sample size of 112 patients. Further studies are needed to evaluate the relationship between traumatic LP and PDPH.

  1. Conclusions

In conclusion, as the RBC count increased in the CSF, the incidence of PDPH decreased among patients with primary headaches. Subse- quently, the admission rate after revisiting our ED or outpatient clinic decreased. Notably, our results suggest that traumatic LP may be an un- expected factor in reducing the occurrence rate of PDPH. Consequently, the admission rate for PDPH was significantly reduced among patients with traumatic LP and those with primary headaches.

Funding

This research did not receive any specific grant from funding agen- cies in the public, commercial, or not-for-profit sectors.

CRediT authorship contribution statement

Juho An: Writing – original draft, Investigation, Data curation. Sung- Eun Lee: Supervision, Methodology. Yura Ko: Writing – review & editing, Supervision, Investigation, Formal analysis, Data curation. Heewon Yang: Writing – review & editing, Writing – original draft, Visualization, Supervision, Methodology, Investigation, Formal analysis, Conceptualization.

Declaration of Competing Interest

None.

Acknowledgments

None.

References

  1. Arevalo-Rodriguez I, Ciapponi A, Roque i Figuls M, Munoz L, Bonfill Cosp X. Posture and fluids for preventing post-dural puncture headache. Cochrane Database Syst Rev. 2016;3:Cd009199.
  2. Grande PO. Mechanisms behind postspinal headache and brain stem compression following lumbar dural puncture–a physiological approach. Acta Anaesthesiol Scand. 2005;49:619-26.
  3. Al-Hashel J, Rady A, Massoud F, Ismail II. Post-dural puncture headache: a prospec- tive study on incidence, risk factors, and clinical characterization of 285 consecutive procedures. BMC Neurol. 2022;22:261.
  4. Amorim JA, Valenca MM. Postdural puncture headache is a risk factor for new postdural puncture headache. Cephalalgia. 2008;28:5-8.
  5. Williams C, Fost N. Doctors’ experience and traumatic Lumbar punctures. Lancet. 1994;344:1086-7.
  6. Shah KH, Richard KM, Nicholas S, Edlow JA. Incidence of traumatic lumbar puncture. Acad Emerg Med. 2003;10:151-4.
  7. Zetlaoui PJ, Buchheit T, Benhamou D. Epidural blood patch: a narrative review. Anaesth Crit Care Pain Med. 2022;41:101138.
  8. Emory ML. Certain reactions following spinal puncture; a comparison of technics. Am Pract Dig Treat. 1948;2:451-6.
  9. Turnbull DK, Shepherd DB. Post-dural puncture headache: pathogenesis, prevention and treatment. Br J Anaesth. 2003;91:718-29.
  10. Rizzoli P, Mullally WJ. Headache. Am J Med. 2018;131:17-24.
  11. Perry JJ, Alyahya B, Sivilotti ML, Bullard MJ, Emond M, Sutherland J, et al. Differenti- ation between traumatic tap and Aneurysmal subarachnoid hemorrhage: prospec- tive cohort study. BMJ. 2015;350:h568.
  12. Cognat E, Koehl B, Lilamand M, Goutagny S, Belbachir A, de Charentenay L, et al. Preventing post-lumbar puncture headache. Ann Emerg Med. 2021;78:443-50.
  13. Bezov D, Ashina S, Lipton R. Post-dural puncture headache: Part II-prevention, man- agement, and prognosis. Headache. 2010;50:1482-98.
  14. Kwak KH. Postdural puncture headache. Korean J Anesthesiol. 2017;70:136-43.
  15. Nath S, Koziarz A, Badhiwala JH, Alhazzani W, Jaeschke R, Sharma S, et al. Atraumatic versus conventional lumbar puncture needles: a systematic review and meta- analysis. Lancet. 2018;391:1197-204.
  16. Arevalo-Rodriguez I, Munoz L, Godoy-Casasbuenas N, Ciapponi A, Arevalo JJ, Boogaard S, et al. Needle gauge and tip designs for preventing post-dural puncture headache (PDPH). Cochrane Database Syst Rev. 2017;4:Cd010807.
  17. Sakurai K, Kanoto M, Nakagawa M, Shimohira M, Tokumaru AM, Kameyama M, et al. Dinosaur tail sign: a useful spinal MRI finding indicative of cerebrospinal fluid leak- age. Headache. 2017;57:917-25.
  18. Leibold RA, Yealy DM, Coppola M, Cantees KK. Post-dural-puncture headache: characteristics, management, and prevention. Ann Emerg Med. 1993;22:1863-70.