Novel pulsatile cerebrospinal fluid model to assess pressure manometry and fluid sampling through spinal needles of different gauge: Support for the use of a 22 G spinal needle with a tapered 27 G pencil-point tip

Abstract

Parallel-walled spinal needles ≤ 22 G are routinely used for lumbar puncture, despite a reported ≥ 32% incidence of post-dural puncture headache. A tapered spinal needle (22 G shaft, 27 G tip) is in use in our institution. We hypothesized that despite the smaller dural puncture hole, this needle has similar cerebrospinal fluid (CSF) pressure equilibration times and CSF sampling times to a standard 22 G needle and assessed a range of spinal needles using an experimental pulsatile CSF reservoir.
The pulsatile CSF reservoir had an oscillating pressure varying between 25 and 15 cm H(2)O at a cycle frequency of 80 s(-1). We tested seven parallel-walled spinal needles (18-27 G) and the tapered 22/27 G needle. CSF pressure was measured every 2 s by manometry. The time to collect 1 ml CSF samples was measured. Saline 0.9% and mannitol 20% were tested separately. One-way ANOVA with Bonferroni post-hoc test was used to compare 22G, 27G and 22/27G needles.
The mean [standard deviation (sd)] CSF pressure equilibration time (saline) was 40.7 (6.4), 108.7 (6.1), and 51.3 (4.6) s for the 22, 27, and 22/27 G needles (P< 0.0001 for comparisons between 27 G and other needles). The mean (sd) CSF sampling time (saline) was 40.3 (3.1), 225.3 (10.0), and 63.0 (5.2) s for the 22, 27, and 22/27 G needles (P< 0.0001 for comparisons between 27 G and other needles, and P= 0.019 between 22 and 22/27 G needles). Saline was different from mannitol for both measurements and all needles (P< 0.0001).
A 22/27 G tapered spinal needle has similar flow properties to the 22 G needle, despite a 27 G tip.

Full text

Novel pulsatile cerebrospinal fluid model to assess pressure
manometry and fluid sampling through spinal needles of
different gauge: support for the use of a 22 G spinal needle
with a tapered 27 G pencil-point tip
Y. Ginosar1*,Y.Smith
3, T. Ben-Hur2, J. M. Lovett4, T. Clements5, Y. D. Ginosar5and E. M. Davidson1
1
Department of Anesthesiology and Critical Care Medicine and
2
Department of Neurology, Hadassah Hebrew University Medical Center,
Jerusalem, Israel
3
Genomic Data Analysis Unit, Hebrew University—Hadassah Medical School, Jerusalem, Israel
4
Undergraduate, Queens College, Queens, NY, USA
5
High School Science Students
* Corresponding author. E-mail: yginosar@netvision.net.il
Editor’s key points
†Post-dural puncture
headache (PDPH) is a
significant problem with
larger bore spinal
needles.
†This study investigated
the flow characteristics of
a tapered spinal needle
(shaft 22 G and tip 27 G)
using saline and
mannitol.
†The flow characteristics
were similar to a
conventional 22 G spinal
needle.
†This tapered needle may
be associated with a
lower incidence of PDPH.
Background. Parallel-walled spinal needles ≤22 G are routinely used for lumbar puncture,
despite a reported ≥32% incidence of post-dural puncture headache. A tapered spinal
needle (22 G shaft, 27 G tip) is in use in our institution. We hypothesized that despite the
smaller dural puncture hole, this needle has similar cerebrospinal fluid (CSF) pressure
equilibration times and CSF sampling times to a standard 22 G needle and assessed a
range of spinal needles using an experimental pulsatile CSF reservoir.
Methods. The pulsatile CSF reservoir had an oscillating pressure varying between 25 and 15
cm H
2
O at a cycle frequency of 80 s
21
. We tested seven parallel-walled spinal needles
(18–27 G) and the tapered 22/27 G needle. CSF pressure was measured every 2 s by
manometry. The time to collect 1 ml CSF samples was measured. Saline 0.9% and
mannitol 20% were tested separately. One-way ANOVA with Bonferroni post-hoc test was
used to compare 22G, 27G and 22/27G needles.
Results. The mean [standard deviation (SD)] CSF pressure equilibration time (saline) was 40.7
(6.4), 108.7 (6.1), and 51.3 (4.6) s for the 22, 27, and 22/27 G needles (P,0.0001 for
comparisons between 27 G and other needles). The mean (SD) CSF sampling time (saline)
was 40.3 (3.1), 225.3 (10.0), and 63.0 (5.2) s for the 22, 27, and 22/27 G needles
(P,0.0001 for comparisons between 27 G and other needles, and P¼0.019 between 22
and 22/27 G needles). Saline was different from mannitol for both measurements and all
needles (P,0.0001).
Conclusions. A 22/27 G tapered spinal needle has similar flow properties to the 22 G needle,
despite a 27 G tip.
Keywords: cerebrospinal fluid; lumbar puncture; post-dural puncture headache
Accepted for publication: 22 September 2011
Dural puncture is one of the most common invasive proce-
dures performed in clinical medicine. Post-dural puncture
headache (PDPH) is a common and distressing complication,
is characterized by severe postural headache, and may be
associated also with nausea and vomiting, neck and back
pain, tinnitus, and rarely diplopia.
1–3
Symptoms, which are
frequently immobilizing, may last from a few days to a few
weeks and occasionally for up to a year.
2
Rare severe compli-
cations include subdural haematoma and death.
4
Iatrogenic
neurological symptoms occurring after diagnostic lumbar
puncture may become superimposed on the presenting
neurological complaints and may confound clinical diagnosis.
Although the pathogenesis of PDPH is not entirely under-
stood, it is associated with the leak of cerebrospinal fluid
(CSF) from the dural puncture site.
5
Accordingly, strategies
to prevent PDPH have focused on technical modifications to
spinal needles aimed at reducing the size of the dural perfor-
ation.
67
There is strong evidence that both the use of narrow
gauge spinal needles
8–15
and the use of pencil-point ‘atrau-
matic’ needle tips
81015–23
reduce the incidence of PDPH
after dural puncture. These are now Type A recommenda-
tions of the American Academy of Neurology.
910
Dural puncture is most commonly performed by anaes-
thetists for the provision of spinal anaesthesia and by
British Journal of Anaesthesia 107 (2): 308–15 (2012)
Advance Access publication 11 December 2011 .doi:10.1093/bja/aer372
&The Author [2011]. Published by Oxford University Press on behalf of the British Journal of Anaesthesia. All rights reserved.
For Permissions, please email: journals.permissions@oup.com

neurologists or paediatricians for diagnostic lumbar puncture.
However, the practice of these groups varies widely.
24–26
Anaesthetists typically use narrow-bore (25–27 G) pencil-
point (‘atraumatic’) spinal needles,
672526
while diagnostic
lumbar puncture is still typically performed using large-bore
(20– 22 G) spinal needles, frequently with cutting or Quincke
tips.
24–26
Not surprisingly, the incidence of PDPH after diag-
nostic lumbar puncture is 32–54%
9102027
compared with
1–2% after spinal anaesthesia.
25
Reasons given for continuing
to use large-bore spinal needles for diagnostic lumbar punc-
ture include the shorter time required to equilibrate the CSF
pressure when using a manometer and the increased speed
in obtaining CSF samples by passive flow.
91028
In an experimental study, Carson and Serpell
28
used an
artificial non-pulsatile CSF reservoir to compare different
spinal needles for the passive flow rate of CSF and for the
time to measure equilibrium CSF pressure by manometry.
They concluded that needles smaller than 22 G are not suit-
able for the measurement of CSF pressure and that 20 G
needles are the needles of choice if CSF sampling is to be
made.
28
However, their model used a non-pulsatile reservoir.
CSF transmits a pulsatile pressure wave, which would be
expected to increase CSF flow from a rigid spinal needle.
29
We designed a pulsatile CSF model in order to test a spinal
needle in use in our institution, which has a wide-bore 22 G
shaft with a tapered 27 G pencil-point tip (Temena-
Polymedic, Temena International, Carriere-sur-Seine,
France).
30
We tested this needle against a range of spinal
needles of different gauge and in two fluids of different vis-
cosity. We hypothesized that in our pulsatile CSF model,
such a needle may improve conditions for lumbar puncture;
the narrow tip minimizing the incidence of PDPH and the
wide shaft reducing the time required to measure CSF pres-
sure manometry and obtain CSF samples.
Methods
We designed a pulsatile CSF model (Fig. 1). A 500 ml fluid bag
with two entry ports was placed inside an inflatable pressure
chamber. The inflating bulb was removed from the pressure
chamber and the hose attached via a tracheal tube connect-
or to a Siemens 908C ventilator. The pressure bag was
inflated using the following ventilator paradigm: pressure
control ventilation 25 cm H
2
O, rate 80 min
21
, PEEP 15 cm
H
2
O, I:Eratio 1:2, horizontal waveform mode). This created
a pulsatile CSF reservoir with a ‘systolic’ and ‘diastolic’ pres-
sure of about 25/15 cm H
2
O (mean CSF pressure 19 cm
H
2
O). The reference port was connected to an optical pres-
sure transducer via a 14 G 3.81 cm long needle. The sampling
CSF reservoir: fluid bag filled
with saline or mannitol 20%
Siemens 900c
ventilator
Pressure control
ventilation:
25/15 cm H2O
Traditional manometer,
with scale gradated every
0.25 cm
Base of manometer fixed
level with CSF reservoir
(not as shown)
Readings made every 2 s
80 cycles min–1
Digital pressure recorder
CSF pressure
measured by
optical pressure
transducer via
reference needle
Reference needle
Spinal needle
3-wa
y
stopcock
Pressure bag
connected to
ventilator Y-piece to
create pulsatile
CSF reservoir
Fig 1 Illustration of the experimental design of the pulsatile CSF reservoir. See text for details.
Tapered 22/27 G spinal needle: CSF flow study BJA
309

port was used to receive the different spinal needles tested in
this study.
CSF pressure was measured via the spinal needle by trad-
itional manometry, using a disposable plastic manometer
(Unomedical A/S DK, Denmark), graduated for the purposes
of this study at 0.25 cm intervals. The manometer was
fixed at the same vertical height as the CSF reservoir.
Before measurement, the stopcock was opened to room air
to allow pressure in the manometer to equilibrate to
ambient pressure (by the escape of fluid). At time 0, the stop-
cock was turned connecting the spinal needle to the man-
ometer. Manometer pressure was measured visually every
2 s for at least 120 s, or longer if the pressure was still
rising. Manometer pressure was read with a resolution of
+0.25 cm; data were measured in triplicate for each
needle for both fluids.
After the measurement of CSF pressure, we measured the
time to obtain 1 ml fluid by passive flow. The spinal needle
was disconnected from the manometer and opened to air.
A 1 ml glass vial with 0.1 ml graduations was held under-
neath the hub of the spinal needle. The time to fill 1 ml
was measured (using a stopwatch) in triplicate for each
needle for both fluids.
We used a range of spinal needles from 18 to 27 G
(Table 1). The primary test needle (22/27 G) had a wide-bore
22 G shaft with a tapered 27 G pencil-point tip (Fig. 2). The
order of assessment of needles was randomized by
computer-generated randomization. Spinal needles and
fluids were concealed and numbered and data were recorded
by a blinded observer.
Reservoir fluid was at room temperature; the thermostat
was set to 208C. Two fluids were used in the pulsatile CSF res-
ervoir: 0.9% saline (Teva Pharmaceuticals Industries, Petach
Tikva, Israel) and 20% mannitol (Baxter Healthcare,
Deerfield, IL, USA). The viscosity and density of 0.9% saline
(at 208C) is 1.002 mPa s
21
(Information from Teva Pharma-
ceutical Industries) and 1.005–1.006 g ml
21
(Information
from Teva Pharmaceutical Industries), respectively; the
viscosity and density of 20% mannitol (at 208C) is 1.5553
mPa s
2131
and 1.07 g ml
21
(Information from Baxter Health-
care), respectively. The viscosity and density of CSF (at 378C)
have been reported as 0.7– 1.0 mPa s
2132
and 1.0003–
1.0007 g ml
21
,
33
respectively.
Statistical analysis
All data were recorded in triplicate and are reported as mean
[standard deviation (SD)] and 95% confidence intervals (CIs)
where appropriate. The measurement of CSF pressure
against time was presented graphically for all needles
tested. In order to avoid multiple statistical tests, only
direct comparisons between the 22, 27, and 22/27 G spinal
needles and between the saline and mannitol used in the
CSF model were assessed using inferential statistics. We
assessed the repeated assessments of CSF pressure over
time by repeated-measures analysis of variance (RM-ANOVA),
using the polynomial contrast function. The between-subject
variables were the fluid in the CSF reservoir (saline vs mannitol)
and the spinal needle used. Mauchly’s test for sphericity was
used to assess the data and if sphericity assumptions could
not be justified, the conservative Greenhouse– Geisser test
was used. Bonferroni’s post hoc test was used to distinguish
between the 22, 27, and 22/27 G spinal needles.We assessed
the time to final CSF pressure equilibration and the time to
obtain a 1 ml sample using one-way ANOVA. Bonferroni’s
post-hoc test was used to distinguish between the 22G,
27G and 22/27G spinal needles. Statistic analyses were per-
formed using SPSS version 17.0, SPSS Inc., Chicago, IL, USA.
Results
All experiments were performed using the same ventilator
settings applied to the inflatable pressure bag enclosing
the CSF fluid reservoir. All needles were tested and there
were no missing data points. The recorded CSF pressure
over time using the manometer connected to the different
spinal needles is shown in Figure 3A(saline) and B(mannitol).
Table 1 The length, gauge (external diameter), internal diameter, needle tip and manufacturer details for each needle studied. Data provided
by Becton–Dickinson, Franklin Lakes, NJ, USA, and Temena-Polymedic (Temena International)
Gauge Manufacturer Tip Internal diameter (mm) External diameter (mm) Length (mm)
18 Becton–Dickinson Quincke 0.864 1.270 88.9
20 Becton–Dickinson Quincke 0.623 0.909 88.9
22 Becton–Dickinson Quincke 0.432 0.719 88.9
24 Temena- Polymedic Sprotte 0.300 0.550 103.0
25 Becton–Dickinson Whitacre 0.279 0.516 88.9
25 Temena- Polymedic Sprotte 0.300 0.500 103.0
26 Becton–Dickinson Whitacre 0.279 0.465 88.9
27 Becton–Dickinson Whitacre 0.279 0.426 88.9
22/27 Temena-Polymedic Sprotte Shaft 0.54 Shaft 0.68 103.0
Tip 0.19 Tip 0.40
BJA Ginosar et al.
310

For clarity, the data for the 22/27 G spinal needle have not
been represented in Figure 3Aand Bbut are compared graph-
ically with the 22 and 27 G spinal needles in Figure 3Cand D.
Comparing the change in CSF pressure measurements over
time with RM-ANOVA, between-subject variables were statistic-
ally significant for both fluid type (P,0.0001) and needle
gauge (P,0.0001). Bonferroni’s post hoc tests were statistic-
ally significant for the comparison between the 22 and 27 G
needles (P,0.0001) and the 22/27 and 27 G needles
(P,0.0001) but not for the comparison between the 22
and 22/27 G needles (P¼0.06).
The time taken until the manometer was able to record
equilibrium ‘CSF pressure’ increased with increasing spinal
needle gauge and with increasing spinal fluid viscosity
(Fig. 4). The mean (SD) time to final equilibrium ‘CSF pressure’
with saline was 40.7 (6.4) s (95% CI 25– 57), 108.7 (6.1) s
(95% CI 93–124), and 51.3 (4.6) s (95% CI 40–63) for the
22, 27, and 22/27 G needles, respectively. The mean (SD)
time to equilibrium ‘CSF pressure’ with mannitol was 62.0
(15.6) s (95% CI 23–101), 118.0 (2.0) s (95% CI 113 –123),
and 61.3 (7.0) s (95% CI 44–79) for the 22, 27, and 22/27
G needles, respectively.
Comparing the time to final CSF pressure equilibration
with one-way ANOVA, there was a significant effect of the
choice of fluid in the CSF reservoir (P,0.0001). For both
fluids, there was approximately a two- to three-fold differ-
ence between the 22 vs 27 G needles (saline P,0.0001,
mannitol P¼0.001) and the 22/27 vs 27 G needles (saline
P,0.0001, mannitol P¼0.001). There was no significant dif-
ference between the 22 vs 22/27 G needles for either fluid
(saline P¼0.019, mannitol P¼1.0).
The time taken to collect 1 ml of ‘CSF’ by passive flow also
increased with increasing spinal needle gauge and with
increasing spinal fluid viscosity (Fig. 5). The time taken to
collect 1 ml of saline was: 40.3 (3.1) s (95% CI 33–48),
225.3 (10.0) s (95% CI 200 – 250), and 63.0 (5.2) s (95% CI
50–76) for the 22, 27, and 22/27 G needles, respectively.
The corresponding times taken to collect 1 ml of mannitol
were 65.3 (3.1) s (95% CI 58–73), 368.0 (8.0) s (95% CI
348–388), and 103.0 (18.4) s (95% CI 57–149). Comparing
the time to collect a 1 ml sample of CSF with one-way
ANOVA, there was a significant effect of the choice of fluid in
the CSF reservoir (P,0.0001). For both fluids, there was a
three- to six-fold difference between the 22 vs 27 G
needles (both fluids P,0.0001) and the 22/27 vs 27 G
needles (both fluids P,0.0001). There was also a significant
difference between the 22 vs 22/27 G needles for both fluids
(saline P¼0.019, mannitol P¼0.023), although this differ-
ence was of smaller magnitude than for the other
comparisons.
Discussion
In this study, we used a pulsatile CSF model to demonstrate
that a wide-bore 22 G shaft spinal needle with a tapered 27 G
pencil-point tip had CSF flow properties similar to those
demonstrated by the 22 G Quincke needle. The equilibrium
time for CSF manometry and the time to obtain a 1 ml
sample were not different for the two needles, even in the
presence of viscous CSF. The 22/27 G spinal needle therefore
should be as practical to use as the 22 G spinal needle
(in terms of the time required for CSF pressure manometry
side orifice, pencil-point tip
27 G
tip
1 mm
103mm
Introducer
needle
Spinal needle
22 G shaft
22 G shaft
ID: 0.54 mm
ED: 0.68 mm
27 G tip
14 mm
ID: 0.19 mm
ED: 0.40 mm
Fig 2 The design of the 22/27 G spinal needle tested in this study. The needle is inserted with the aid of an external introducer needle. The
spinal needle is 103 mm in length; the proximal 88 mm has a 22 G external diameter and the distal 14 mm has a 27 G external diameter with a
pencil-point tip. Line drawing modified from figure received from Temena International and reproduced with permission.
Tapered 22/27 G spinal needle: CSF flow study BJA
311

and CSF sampling), but with a lower incidence and severity of
PDPH. This was in contrast to the regular 27 G spinal needle,
where the mean time required to measure CSF pressure by
manometry (108 s) and the mean time required to obtain
a 1 ml CSF sample (225 s) were both too long for routine clin-
ical diagnostic use.
CSF is a pulsatile fluid and accordingly we chose to use a
pulsatile CSF model in this study. Previous studies have used
non-pulsatile, artificial CSF chambers to assess flow rates
through spinal needles of different design,
28 34 –37
either to
determine the time until CSF flashback
34 –37
(an important
landmark confirming intrathecal placement) or for CSF sam-
pling time.
28
Unlike pulsatile flow in an elastic-walled tube
(e.g. aorta), spinal needles are rigid non-elastic tubes. The
physical properties of pulsatile flow in a rigid tube have
been described.
29
In a rigid tube, systolic peak pressures
are transmitted instantaneously through the needle
without being absorbed. These higher systolic pressure
peaks cause an additional ‘squirt’ of CSF to be expelled
from the needle at each systole. In non-pulsatile flow,
where flow is driven by only a static pressure, this does not
happen. As a consequence, flow is expected to be greater
in the pulsatile state and the use of a non-pulsatile CSF
chamber may significantly under-estimate the passive flow
of CSF through spinal needles.
There are several limitations to this study. We did not use
a range of different CSF pressures and the pressure chosen
was moderately high (mean 19 cm H
2
O), compared with
12–15 cm H
2
O normally encountered with the patient lying
in the lateral position. Increased CSF pressure will increase
the passive flow of CSF and reduce the time to collect CSF
samples. This potential limitation does not diminish the rela-
tive difference observed in CSF sampling times between dif-
ferent spinal needles. Importantly, the slightly increased
0
0
5
10
CSF pressure (cm H2O)
15
20
25
20 40 60
Time (s)
Saline 0.9% Saline 0.9%
Mannitol 20% Mannitol 20%
80 100 120 0
5
10
CSF pressure (cm H2O)
15
20
25
200
00
40 60
Time (s)
80 100 120
0
5
10
CSF pressure (cm H2O)
15
20
25
20 40 60
Time (s)
80 100 120
0
5
10
CSF pressure (cm H2O)
15
20
25
20 40 60
Time (s)
80 100 120
18 G
20 G
22 G
24 G
25 G W
25 G Poly
26 G
27 G
22 G
27 G
22/27 G
*
*
18 G
20 G
22 G
24 G
25 G W
25 G Poly
26 G
27 G
22 G
27 G
22/27 G
A
B
C
D
Fig 3 CSF pressure measured by manometry every 2 s for 120 s for a range of spinal needles (see Table 1for details of needles). The equili-
bration time for the spinal needles increased as the internal diameter decreased. The equilibration time increased with increasing fluid viscos-
ity, as can be seen by comparing 0.9% sodium chloride (Aand C)vs 20% mannitol (Band D). (Aand B) The CSF manometry pressure over time for
all spinal needles assessed in this study. Coloured lines represent mean data; SDs have been omitted for the sake of clarity from these graphs.
(Cand D) The CSF manometry pressure over time for the three spinal needles directly compared in this study: 22, 27, and the tapered 22/27 G
spinal needle (22 G shaft, 27 G pencil-point tip). Data are presented as mean (SD) based on triplicate measures for each spinal needle. The
change in CSF pressure measurements over time for both the 22 and 22/27 G needles was significantly different from the 27 G needle
(RM-ANOVA,*P,0.0001) but not significantly different from each other.
BJA Ginosar et al.
312

CSF pressure in our set-up should not have affected the time
to equilibrium CSF pressure by manometry as the increased
rate of filling the manometry tubing is exactly offset by the
increased vertical height that needs to be filled. Another limi-
tation was that the CSF was maintained at room tempera-
ture, and as a consequence, the fluids used had a slightly
higher viscosity than would have been observed at body tem-
perature.
38
These limitations partly offset each other as
regards possible effects on the time to collection of CSF
samples. Nevertheless, the main observations are not
affected by these concerns, specifically (i) the time to
measure equilibrium CSF pressure and the time to collect a
1 ml CSF sample are profoundly affected by the gauge of
the spinal needle and to a lesser degree by the viscosity of
the CSF fluid and (ii) the flow properties of the 22/27 G
needle are similar to those of the 22 G needle.
A further potential limitation is that the needles tested
differed in respects other than gauge, in particular the
needle tip orifice area and the length of the needle
(Table 1). However, when compared with gauge, the effects
of both orifice area and needle length are relatively minor.
6
The lack of effect of orifice area was illustrated clearly in a
published study where the area of the side orifice was
reduced from 1.7 ×0.32 to 0.32 ×0.32 mm with no effect
on observed flow rates.
39
In that study, based on both
theoretical calculations and observational data, it was
demonstrated that only by reducing the orifice size below
the cross-sectional area of the needle would flow rates be
affected. None of the needles used in our study had orifice
sizes smaller than needle cross-sectional area. Regarding
needle length, from Poiseuille’s law, flow rates are propor-
tional to the fourth power of needle radius and inversely pro-
portional to only the first power of needle length. In
turbulent flow, the dependence on needle radius is even
greater (fifth power of radius). Clearly, when comparing the
27 and 22/27 G needles, the 14.1 mm increase in length of
the 22/27 G needle would be expected to reduce flow rates
to some degree, but this would be expected to be dwarfed
by the effect due to increased shaft radius. Thus, it may
not be surprising that despite the different needle lengths
and orifice designs of the needles in this study, it was the
diameter of the needle shaft that seemed to have the over-
whelming impact. The 22/27 G needle had flow properties
between those of the 22 and 24 G needles and markedly dif-
ferent from those of the 27 G needle. Had the 22/27 G needle
been of the same length as the 22 G needle used, it is likely
that the similarity in flow to the 22 G needle would have been
even closer.
There are two conflicting demands on the design of a spinal
needle. With reducing spinal needle gauge (increasing needle
diameter), the rate of CSF flashback, the CSF sampling rate,
and the rate of reaching equilibrium CSF pressure by manome-
try are all increased. A 22 G needle is probably the smallest
18
0
20
40
60
80
100
120
19 20
Saline 0.9%
Mannitol 20%
21 22 23
Spinal needle gauge
Time to measure CSF pressure (s)
24 25 26 27
**
*
Fig 4 The time to measure equilibrated CSF pressure for a range
of spinal needles (see Table 1for details of needles). The equili-
bration times increased with decreasing internal diameter and
with increasing fluid viscosity. The square symbols represent
the 22/27 G spinal needle, with a 27 G tip and 22 G shaft; CSF
sample collection times through this needle were not different
from the traditional 22 G spinal needle. Means and SDs based
on triplicate measures for each spinal needle. Significant differ-
ence for final equilibration times (one-way ANOVA with Bonferroni)
for both 27 vs 22 G and 27 vs 22/27 G in saline (*P,0.0001) and
mannitol (**P¼0.001). There was no significant difference
between the 22 and 22/27 G needles for either fluid.
18
0
50
100
150
200
250
300
350
400
19 20
Saline 0.9%
Mannitol 20%
21 22 23
Spinal needle gauge
Time to collection 1 ml sample (s)
24 25 26 27
**
*
Fig 5 The time to collect a 1 ml CSF sample for a range of spinal
needles (see Table 1for details of needles). The sample collection
time increased with decreasing internal diameter and with
increasing fluid viscosity. The square symbols represent the 22/
27 G spinal needle, with a 27 G tip and 22 G shaft; CSF sample
collection times through this needle were not different from
the traditional 22 G spinal needle. Means and SDs based on trip-
licate measures for each spinal needle. Significant difference
for final equilibration times (one-way ANOVA with Bonferroni) for
both 27 vs 22 and 27 vs 22/27 G in saline (*P,0.0001) and man-
nitol (**P¼0.001). There was a smaller magnitude but significant
difference between the 22 and 22/27 G needles for both saline
(P¼0.019) and mannitol (P¼0.023).
Tapered 22/27 G spinal needle: CSF flow study BJA
313

needle compatible with the flow rates needed for routine diag-
nostic lumbar puncture.
928
On the other hand, with increasing
spinal needle gauge, the incidence of PDPH reduces. For 22 G
spinal needles, the PDPH rate is 25–54%
910202227
for
cutting needles and 7–8% for pencil-point needles.
22
These
rates compare with reported PDPH rates for 25 G needles:
8.5%
21
to 17.9%
23
(cutting) and 3% (pencil point)
21
; for 26 G
needles: 9.6% (cutting)
14
and 2.7% (pencil point);
14
and for
27 G needles: 1.5–1.7%.
6 8 14 15
The combination of a 27 G
pencil-point tip and a wide-bore 22 G shaft may be an ideal
combination for a spinal needle for diagnostic lumbar punc-
ture. There were minor differences in flow properties
between the 22/27 and 22 G spinal needles, particularly a 22
s increase in the time to collect a 1 ml saline sample.
However, it may be questioned whether this would
justify the higher PDPH rates associated with a 27 G dural
puncture hole.
We have used over 30 000 of these 22/27 G tapered spinal
needles over the past 14 yr in our institution for the provision
of spinal anaesthesia and analgesia in a mixed surgical and
obstetric population. Most of these anaesthetics were admi-
nistered by junior residents and no serious complications
have occurred (with the exception of one needle that frac-
tured in subcutaneous tissue requiring removal under fluor-
oscopy). The incidence of PDPH is below 1% in our obstetric
population in patients receiving the 22/27 G tapered spinal
needle. Although the learning curve may be longer and the
cost significantly greater for 22/27 G tapered spinal needles
when compared with conventional 22 G spinal needles,
these are probably justified if they can reduce the incidence
of an immobilizing headache after diagnostic lumbar punc-
ture from 25– 54%
910202227
(or even 7–8%
22
) down to
1.5–1.7%.
681415
As the authors of the American Academy of Neurology
report wrote,
9
quoting Tourtellotte and colleagues
40
(writing in the neurology literature nearly 50 yr ago): ‘If
patients undergoing an LP for diagnostic purposes were
treated like patients undergoing spinal anaesthesia, the fre-
quency of PDPH could be markedly reduced’.
In summary, this study describes the use of a pulsatile CSF
model to assess the equilibration time for pressure manome-
try and CSF sampling rates through a range of spinal needles
of different gauge and with fluids of different viscosity. A 22/
27 G tapered spinal needle (with a 22 G shaft and a narrow
27 G pencil-point tip) has flow properties similar to those
of the 22 G spinal needle but creates a 27 G rather than a
22 G puncture hole in the dura. This would be expected to
reduce the incidence and severity of PDPH, typically asso-
ciated with large-bore needles.
Acknowledgements
The authors gratefully acknowledge the assistance of the fol-
lowing: Prof. Mair Zamir, Emeritus Professor, Department of
Applied Mathematics, University of Western Ontario,
Canada, for explaining the physics of pulsatile flow in a
rigid vs non-rigid conduit; Alex Tuchband for technical
assistance with the pulsatile CSF model; Alex Tuchband and
Orit Amsallem for helping to track down manufacturer speci-
fications; Becton–Dickinson and Temena International for
providing technical information regarding the internal and
external diameters of the spinal needles; Baxter Healthcare
and Teva Pharmaceutical Industries for technical information
regarding the density, viscosity, or both of the fluids used in
the CSF model; Temena International for providing a line
drawing of their needle, a modified version of which was
reproduced with their permission in Figure 2. Y.G. designed
the study, constructed the CSF model, performed the statis-
tics, and wrote the manuscript. He had full access to all of
the data in the study and takes responsibility for the integrity
of the data and the accuracy of the data analysis. Y.S. per-
formed modelling on the data (not reported) and checked
the statistics. T.B.-H. provided valuable clinical insights and
edited the manuscript. Y.D.G., T.C. and J.M.L. collected and
analysed data (under supervision from E.M.D. and Y.G.) as
part of their summer science project. E.M.D. supervised
data collection, provided valuable clinical insights, and
edited the manuscript.
Declaration of interest
None declared.
References
1 Ahmed SV, Jayawama C, Jude E. Post lumbar puncture headache:
diagnosis and management. Postgrad Med J 2006; 82: 713– 6
2 Lybecker H, Dhernes M, Schmidt JF. Post-dural puncture headache
(PDPH): onset, duration, severity and associated symptoms. An
analysis of 75 consecutive patients with PDPH. Acta Anesthesiol
Scand 1995; 39: 605– 12
3 Vilming ST, Kloster R. Pain location and associated symptoms in
postlumbar puncture headache. Cephalalgia 1998; 18: 697–703
4 Newrick P, Read D. Subdural haematoma as a complication of
spinal anesthetic. Br Med J 1982; 285: 341– 2
5 Turnbull DK, Shepherd DB. Post-dural puncture headache: patho-
genesis, prevention and treatment. Br J Anaesth 2003; 91:
718– 29
6 Tsen L, Hepner DL. Needles used for spinal anesthesia. Expert Rev
Med Devices 2006; 3:1–10
7 Calthorpe N. The history of spinal needles: getting to the point.
Anaesthesia 2004; 59: 1231– 41
8 Halpern S, Preston R. Post-dural puncture headache and spinal
needle design. Meta-analyses. Anesthesiology 1994; 81: 1376– 83
9 Evans RW, Armon C, Frohman EM, Goodin DS. Assessment: pre-
vention of post-lumbar puncture headaches: report of the thera-
peutics and technology assessment subcommittee of the
American Academy of Neurology. Neurology 2000; 55: 909– 14
10 Armon C, Evans RW. Addendum to assessment: prevention of
post-lumbar puncture headaches. Report of the therapeutics
and technology assessment subcommittee of the American
Academy of Neurology. Neurology 2005; 65: 510– 12
11 Tourtellotte WW, Henderson WG, Tucker RP, Gilland O, Walker JE,
Kokman E. A randomized, double-blind clinical trial comparing
the 22 versus 26 gauge needle in the production of the post-
lumbar puncture syndrome in normal individuals. Headache
1972; 12:73–8
BJA Ginosar et al.
314

12 Flaatten H, Rodt SA, Vamnes J, Rosland J, Wisborg T, Koller ME.
Post-dural puncture headache. A comparison between 26- and
29-gauge needles in young patients. Anaesthesia 1989; 44:
147– 9
13 Geurts JW, Haanschoten MC, van Wijk RM, Kraak H, Besse TC.
Post-dural puncture headache in young patients. A comparative
study between the use of 0.52 mm (25-gauge) and 0.33 mm
(29-gauge) spinal needles. Acta Anesthesiol Scand 1990; 34:
350– 3
14 Kang SB, Goodnough DE, Lee YK, et al. Comparison of 26- and
27-G needles for spinal anesthesia for ambulatory surgery
patients. Anesthesiology 1992; 76: 734– 8
15 Choi PT, Galinski SE, Takeuchi L, Lucas S, Tamayo C, Jadad AR.
PDPH is a common complication of neuraxial blockade in parturi-
ents: a meta-analysis of obstetrical studies. Can J Anaesth 2003;
50: 460–9
16 Hart JR, Whitacre RJ. Pencil-point needle in prevention of post-
spinal headache. J Am Med Assoc 1951; 147: 657– 8
17 Sprotte G, Schedel R, Pajunk H. An ‘atraumatic’ universal needle
for single-shot regional anesthesia: clinical results and a 6 year
trial in over 30,000 regional anesthesias [in German]. Reg
Anesth 1987; 10: 104– 8
18 Strupp M, Schueler O, Straube A, Von Stuckrad-Barre S, Brandt T.
‘Atraumatic’ Sprotte needle reduces the incidence of post-lumbar
puncture headaches. Neurology 2001; 57: 2310– 2
19 Lavi R, Yarnitzky D, Rowe JM, Weissman A, Segal D, Avivi I. Stand-
ard vs atraumatic Whitacre needle for diagnostic lumbar punc-
ture: a randomized trial. Neurology 2006; 67: 1492– 4
20 Thomas SR, Jamieson DRS, Muir KW. Randomised, controlled trial
of atraumatic versus standard needles for diagnostic lumbar
puncture. Br Med J 2000; 321: 986– 90
21 Buettner J, Wresch KP, Klose R. Post-dural puncture headache:
comparison of 25-gauge Whitacre and Quincke needles. Reg
Anesth 1993; 18: 166– 9
22 Prager JM, Roychowdhury S, Gorey MT, et al. Spinal headaches
after myelograms: comparison of needle types. Am J Roentgenol
1996; 167: 1289– 92
23 Tarkkila PJ, Heine H, Tervo RR. Comparison of Sprotte and Quincke
needles with respect to post dural puncture headache and back-
ache. Reg Anesth 1992; 17: 283– 7
24 Serpell MG, Haldane GJ, Jamieson DRS, Carson D. Prevention of
headache after lumbar puncture: questionnaire survey of neurol-
ogists and neurosurgeons in United Kingdom. Br Med J 1998; 316:
1709– 10
25 Birnbach DJ, Kuroda MM, Sternman D, Thys DM. Use of atraumatic
spinal needles among neurologists in the United States. Head-
ache 2001; 41: 385– 90
26 Serpell MG, Rawal N. Headaches after diagnostic dural puncture:
smaller, atraumatic needles and protocols for early treatment
should reduce morbidity. Br Med J 2000; 321: 973– 4
27 Kuntz KM, Kokmen E, Stevens JC, Miller P, Offord KP, Ho MM. Post-
lumbar puncture headaches: experience in 501 consecutive pro-
cedures. Neurology 1992; 42: 1884– 7
28 Carson D, Serpell M. Choosing the best needle for diagnostic
lumbar puncture. Neurology 1996; 47:33–7
29 Zamir M. The Physics of Pulsatile Flow. New York, NY: Springer-
Verlag, 2000
30 Fukumitsu K, Sakai T, Yoshikawa K. Reduced-tip needle for spinal
anesthesia. Anesth Analg 1994; 79: 1207– 8
31 Zhu C, Ma Y, Zhou C. Densities and viscosities of sugar alcohol
aqueous solutions. J Chem Eng Data 2010; 55: 3882–5
32 Richardson MG, Wissler RN. Density of lumbar cerebrospinal fluid
in pregnant and nonpregnant humans. Anesthesiology 1996; 85:
326– 30
33 Bloomfield IG, Johnston IH, Bilston LE. Effects of proteins, blood
cells and glucose on the viscosity of cerebrospinal fluid. Pediatr
Neurosurg 1998; 28: 246– 51
34 Abouleish E, Mitchell M, Taylor G, Miller H, Warters D, Rashad MN.
Comparative flow rates of saline in commonly used
spinal needles including pencil-tip needles. Reg Anesth 1994;
19:34–42
35 Gerrish SP, Peacock JE. Variations in the flow of
cerebrospinal fluid through spinal needles. Br J Anaesth 1987;
59: 1465– 71
36 Lyons G, Lesser P, Bembridge M. Lessons from in vitro testing of
spinal needles. Anaesthesia 1995; 50: 964– 8
37 Cruickshank RH, Hopkinson JM. Fluid flow through dural puncture
sites: an in vitro comparison of needle point types. Anaesthesia
1989; 44: 415– 8
38 Heller AR, Zimmerman K, Seele K, Rossel T, Koch T, Litz RJ. Modi-
fying the baricity of local anesthetics for spinal anesthesia by
temperature adjustment: model calculation. Anesthesiology
2006; 105: 346– 53
39 Aglan MY, Stansby PK. Modification to the Sprotte spinal needle.
Anaesthesia 1992; 47: 506– 7
40 Tourtellotte WW, Haerer AF, Heller GL, et al. Post-lumbar Puncture
Headaches. Springfield, IL: Charles C. Thomas, 1964
Tapered 22/27 G spinal needle: CSF flow study BJA
315