Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=iivs20
Journal of Investigative Surgery
ISSN: 0894-1939 (Print) 1521-0553 (Online) Journal homepage: https://www.tandfonline.com/loi/iivs20
Estimating the Lumbar Puncture Needle Depth in
Children
Derya Celik, Ozkan Onal, Seza Apiliogullari, Inci Kara & Jale Bengi Celik
To cite this article: Derya Celik, Ozkan Onal, Seza Apiliogullari, Inci Kara & Jale Bengi Celik (2021) Estimating the Lumbar Puncture Needle Depth in Children, Journal of Investigative Surgery, 34:2, 157-163, DOI: 10.1080/08941939.2019.1614698
To link to this article: https://doi.org/10.1080/08941939.2019.1614698
Published online: 22 May 2019.
Submit your article to this journal
Article views: 69
View related articles
View Crossmark data
ORIGINAL RESEARCH
Estimating the Lumbar Puncture Needle Depth
in Children
Derya Celik, Ozkan Onal, Seza Apiliogullari, Inci Kara, Jale Bengi Celik
Department of Anesthesiology and Intensive Care, Medical Faculty, Selcuk University, Konya, Turkey
ABSTRACT
Background: Lumbar puncture’s (LP) success is dependent on the skill of the physician, anatomy, size, and posture of the patient. Aims: The purpose of this study was to describe a method that could be used to help estimate the correct depth of needle (Y) insertion in children based on age, weight (W), and height (H). Methods: The study consisted of 200 children American Society of Anesthesiologist class I–II aged 0–12 years who underwent spinal block for orthopedic, pediatric, and genitourinary surgery. The distance from the skin entry point to the tip of the spinal needle was measured after the LP was performed. The relationship between the Y and W, H and body mass index (BMI) was calculated. Predictive statistical models were used to determine the LP needle depth. A paired sample t-test was conducted to compare the findings of the developed model with those of earlier models. Results: The patients were aged 2–144 months, with H and W of 43–154 cm and 2.5–48 kg, respectively. The BMI was 10.75–37.72 kg/m2
. Before the Y was esti-mated, the relationship between the independent variables and the depth variable, which was the depend-ent variable, was examined. According to the obtained results, the model consists of strong relationships with H, W, and Hþ W. The formula for predicting Y based on W plus H was as follows: for all patients: Y (cm)¼ 0.861 þ 0.012 H (cm) þ 0.035 W (kg). Based on H, the formula for predicting the required Y was as follows: For all patients: Y (cm) ¼ 0.393 þ 0.023 H (cm). Based on W, the formula for predicting the required Y was as follows: For all patients: Y (cm)¼ 1.460 þ [0.067 W (kg)]. Conclusion: The formula may provide a more reliable estimate of the required LP depth in children than that obtained using current mod-els. However, larger studies are needed to standardize the formula.
Keywords: children; depth; lumbar puncture; spinal anesthesia
INTRODUCTION
A lumbar puncture (LP) has long been used for diagnostic purposes and sometimes treatment in various disciplines. LP is an important diagnostic way for the central nervous system infectious dis-eases such as meningitis [1]. In addition, intrathecal drugs are applied using LP in the treatment of chil-dren with acute lymphoblastic leukemia and menin-geal metastases [2]. Intrathecal opioid use in cancer pains is quite effective in children [3]. The first report of spinal anesthesia was in 1898, when August Bier administered spinal block to six patients, two of whom were children [4]. In 1899, Bainbridge used this technique in a herniotomy
operation in a 3-month old infant [5]. After 1984, the popularity of spinal anesthesia increased, especially among children who may be at risk from general anesthesia [6]. The accuracy of the location of a LP is confirmed by a constant flow of cerebrospinal fluid (CSF) from the LP needle [7].
The spinal cord is at the level of L3 at birth. Between 6 months and 1 year, it reaches the level of L1. The spinal needle should be inserted from the L4-L5 or L5-S1 space. As the spinal colon is more flexible in children than adults, the intervertebral space can be reached more readily [8]. Therefore, an LP can be performed relatively easily in children according to adults. For a LP and spinal anesthesia, children should be placed in a seated or lateral
Received 29 March 2019; accepted 30 April 2019.
Address correspondence to Ozkan Onal, Department of Anesthesiology and Intensive Care, Selcuk University Alaaddin Keykubat Campus, Istanbul Street, Yazir District, 42100, Konya, Turkey. E-mail:drozkanonal@selcuk.edu.tr
Color versions of one or more of the figures in the article can be found online atwww.tandfonline.com/iivs. ISSN: 0894-1939 print / 1521-0553 online
position [9]. The LP injection should be adminis-tered slowly to avoid rapid spread of spinal anesthe-sia [10]. After spinal block, the Trendelenburg position should be avoided, as this can elevate the block and potentially lead to complications, such as severe respiratory arrest [11].
In pediatric patients, spinal anesthesia is pre-ferred due to its rapid onset of action and adequate motor and sensory block [12]. Furthermore, pediatric spinal anesthesia is easy to perform, effective, and safe [5, 6,8,13,14]. As spinal anesthesia is the most effective method to suppress surgical stress and the cardiovascular response in children, some anesthe-tists contend that spinal block should be preferred, even in pediatric cardiac surgery [15]. However, at younger ages, the duration of action of spinal anes-thesia decreases [10].
Spinal anesthesia is a practical alternative to general anesthesia, as it is safe, easy, and cheap [16]. It also has the befit of rapid turnover in the operat-ing theater [16]. In children, to prolong the duration of action of spinal anesthesia in children, various adjuvants may be employed. Those most commonly used for this purpose are adrenaline, opioids, and clonidine [6,10].
A LP is important not only for anesthetists but also for pediatricians, oncologists, and infectious dis-ease experts. Spinal anesthesia may be recom-mended as an alternative to general anesthesia for children, including newborns and adults [17]. It may also be recommended as the first choice. The advan-tages of spinal anesthesia include rapid onset of action, adequate motor and sensory block, no need for tracheal incubation, decreased stress response to surgery, maintenance of analgesia in the post-opera-tive period, decreased need for opioids, reduced bleeding, and protection of pulmonary functions. The most important side effect of spinal anesthesia described in adults is the development of severe hypotension during surgery. However, in children, unlike adults, spinal anesthesia does not lead to hypotension, which is an important additional advantage [10,14,18].
As demonstrated in previous studies, in experi-enced hands, spinal anesthesia is a safe method, even in high-risk groups, such as newborns and children. During LP administration using pencil point spinal needles in adults, a “click” is felt while passing the duramater [12]. In pediatric patients, the distance between the skin and dura may be very short, and the “click” may not be felt. Estimation of the distance to CSF makes it easier to carry out LP.
It is mandatory to carry out studies to investi-gate various methods for successful LP application in pediatric patients. One of the issues that need to be emphasized to increase success is to estimate the depth of the needle. Therefore, guidelines are
needed to improve the success of LP in pediat-ric patients.
So far, there has not been a lot of study on the estimation of needle depth in pediatric patients, and the studies have reached very different results and there is no association on the formulas found [9,19–24].
The aim of this study was to develop a formula for estimating the required Y in LPs using W and H values of children to aid the success and comfort of LPs in children. And our hypothesis is that our model has a higher needle depth estimation value and higher reliability values than Abe et al.’s [9], Bilic et al.’s [19], Chong et al.’s [20], Craig et al.’s
[22], Arthurs et al.’s [23] models.
Our model was compared with five different models accepted in the medical literature and the model we obtained was found to be different from the other five models. The primary outcome of this study is that the model we created is different from the other five accepted models and a new model has been introduced. In addition, this study shows that our model has the highest Y estimation value and highest reliability values according to the other five models.
MATERIAL AND METHODS
After obtaining approval from Selcuk University Clinical Investigations Ethics Committee and informed consent from the parents, 200 children aged 0–12 years American Society of Anesthesiologist (ASA) class I–II scheduled for surgery under spinal anesthesia between 1 February 2012 and 1 January 2013 in the Faculty of Medicine, Selcuk University Hospital were included and this study was per-formed according to Declaration of Helsinki.
Children with contraindications for spinal anes-thesia and ASA> class II were not included in the study. After fasting for 8 h, the children were taken to the operating theater. As standard, electrocardio-gram monitoring was conducted, in addition to heart beat rate, noninvasive systolic blood pressure, diastolic blood pressure, mean blood pressure, and peripheral oxygen saturation monitoring. Materials required for spinal block were prepared under ster-ile conditions. Anesthesia was induced with 6%–7% sevoflurane, 60% nitrous oxide, and 40% oxygen using a face mask.
After the induction of anesthesia, via the intra-venous route, an infusion of 3–7 ml/kg/h of 0.45% NaCI and 5% dextrose fluid was started. The patient’s knees were pulled toward the abdomen, and the patient’s legs were brought to flexion to place the patient in a lateral decubitus position. The area where the block was to be administered was 158 D. Celik et al.
cleaned three times with 10% povidone-iodine from the center to the periphery in a circular manner. The area was then covered with a sterile sponge, leaving the administration point at the middle. The vertebral space was found by palpation of the lumbar verte-bral space (L4-L5 or L5-S1). The midline skin was then entered vertically using a 25 gauge (G) (0.53 88 mm; Spinocan Braun, Melsungen AG, Melsungen, Germany) spinal needle. Successful LP was verified with the observation of CSF flow. Subsequently, 0.5% hyperbaric bupivacaine was administered as local anesthetic. The doses adminis-tered varied according to the age of the patient. For 1-month to 1-year-old patients, the dose was: 0.8 mg/kg, and for patients aged 1–5 years, the dose was 0.5 mg/kg. For patients older than 5 years, the dose was 0.4 mg/kg. The maximum dose did not exceed 10 mg [25]. Before withdrawing the spinal needle, the depth of the needle was marked with a skin pen, and the distance from the skin entry point to the tip of the spinal needle was measured using a ruler. The needle was then withdrawn.
Anesthesia was maintained via a face mask and laryngeal mask or I-gel at the discretion of the anes-thetist. Superficial anesthesia was maintained with sevoflurane, or sevoflurane was discontinued based on hemodynamic data (MAC<1). MAC is useful as a standard of anesthetic potency. After the surgical intervention, the patients were awakened and trans-ferred to the post-operative recovery room, accom-panied by the anesthetist. All patients were monitored for at least for 1 h in the recovery room. When they were hemodynamically stable, conscious, and able to move their feet, they were transferred to the clinic, accompanied by their parents.
The sample size was determined as 200 patients based on a study by Bilic et al. [19] of 195 children aged 1 day to 19 years and for the linear regression to be applied as a ¼ 0.05 (Type I error), b ¼ 0.80 (Power of Test), f2¼ 0:05 (Effect size), the sample
size was calculated as N: 196 in G-Power Program. For estimation of the Y (dependent variable), a formula was developed based on W and H values (independent variables) of children aged 1–12 months and children older than 1 year. To develop a novel method, a simple linear regression model was used, and statistical analyses were car-ried out using IBM SPSS Statistics, version 21. To determine whether the Y varied with level, an inde-pendent sample t-test (Student’s t-test) was per-formed. The final model that was obtained was compared with five accepted models to determine differences between the models of LP, Y estimation and which models were most suitable. In the com-parison of the six models, a paired sample t-test was employed. A p< 0.05 was considered statistically significant.
RESULTS
Two hundred children aged 0–12 years ASA class I-II who were scheduled to undergo surgery under spinal anesthesia in the Faculty of Medicine, Selcuk University Hospital, were included in this study. In all patients, LPs were carried out successfully.
As there were only two children in the newborn group, these patients were excluded from the study, and the statistical evaluation included the remaining 198 patients. Demographic information and descrip-tive statistics according to sex on these 198 patients are presented inTables 1 and2, respectively.
Our model established for Hþ W is significant (p¼ 0.00 < 0.05). The H and W variables in the regression model are also significant. Our regression formulae which is formed by using H and W varia-bles is as follows:
Y cmð Þ ¼ 0:861 þ 0:012 H ð Þ þ 0:035 Wð Þ Using the data obtained for all individuals, the results for five different models were compared. In addition, the model we developed was compared with five models accepted in the medical literature, and our model was found to be different from the five other models. (p¼ 0.00, p < 0.05). The formulae developed to calculate needle depth in the afore-mentioned five models are presented below:
TABLE 1. Demographic characteristics of 198 patients Characteristic Frequency/interval Mean Sex Female 55 Male 143 Age (month) 2–144 42.872.57 Height (cm) 43–154 91.971.76 Weight (kg) 2.50–48 15.120.59 (BMI) 10.75–37.72 16.930.24 Needle depth (mm) 0.9–4.5 2.470.05 Level (n) L4-L5 150 L5-S1 48 Operation type Child surgery 150 Orthopedic surgery 48
TABLE 2. Descriptive statistics according to sex
Sex Characteristic Mean (Min–Max) Male Age (month) 41.98 (2–144)
Height (cm) 191.44 (43–154) Weight (kg) 15.23 (2.50–45) BMI 17.31 (12.35–37.72) Female Age (month) 45.18 (2–144)
Height (cm) 93.34 (50–152) Weight (kg) 14.86 (3–48)
Formula Y cmð Þ ¼ 1 þ 17 W=H 9ð Þ Y cmð Þ ¼ 1:3 þ 0:07 W 20ð Þ Y cmð Þ ¼ 10 W=Hð Þ þ 1 21ð Þ Y cmð Þ ¼ 0:03cm H cmð Þ 23ð Þ Y mmð Þ ¼ 2 W þ 7ð Þ 10 24ð Þ
This shows that the obtained model was different from the five accepted models and that a new model was produced. To determine the difference between the models, a dependent sample test was used. The H and W values obtained in the proposed model were tested in all the models, and an estimation value was found for each model. A comparison of the five accepted models and the model developed herein is shown in Table 3.
As shown in Table 4, a comparison of the Y val-ues in the models showed that the strongest correl-ation was found in the model developed herein. In addition, the new yielded better results than the five accepted models. As shown in Table 5, the sum of the squares of the difference between estimated and observed values gives the PRESS value for a particu-lar model. The PRESS value indicates how well an estimated value of a model predicts an observed dependent variable, with lower PRESS values signi-fying better model estimation. The sum of the square of difference between estimated and observed values yields the most accurate classical linear regression models. The model with the small-est PRESS value is considered the most accurate. According to the results obtained, the most accurate regression model seemed to be the model developed in this study.
Figure 1 shows the scatter plot of the estimated depth values obtained using the model developed herein and those of the other models. As can be seen, the estimated values of Bilic et al. [19], Arthurs
et al. [23], and Abe et al. [9] deviated markedly from actual values and deviated from the regres-sion line.
DISCUSSION
The widespread use of LPs has made it imperative to conduct studies to investigate various methods for successful LP administration. One issue that needs to be addressed to increase LP success is the estimation of Y. Especially in pediatric age groups, successful LP is more important due to their young ages. In the pediatric age group, the intervention should be carried out as rapidly as possible and with minimum traumatization. Guiding information is required to enhance the success of LPs in pediat-ric patients. In our clinic, providing anesthesia with subarachnoid block with LP in a commonly used method of anesthesia induction in selected surgi-cal procedures.
In children, the use of spinal anesthesia has recently become popular for selected surgeries. Its use has become even more widespread among pre-term infants. The popularity of spinal anesthesia is mainly due to severe side effects of general anesthe-sia and sedation, such as apnea, hypoxemia, and bradycardia, in premature infants and other pediat-ric age groups [21].
The aim of this study was to develop a model to estimate LP needle depth based on our experience and compare its performance with that of accepted models in the literature [9, 19, 20, 22, 23]. Success during LP for spinal anesthesia or a different pro-cedure depends not only on the previous experience of the anesthetist and body posture of the patient but also on needle selection. For example, if too short a needle is selected, the needle will not be long enough to reach the desired space. Conversely, if too long a needle is chosen, it will be quite TABLE 3. Comparison of five accepted models with the model developed herein
For model difference 95% confidence interval
Models Needle depth averages Standard deviation Lower limit Upper limit p values Ours 2.4942 0.26525 –0.08628 –0.01193 0.01 Chong’s 2.5433 Ours 2.4942 0.12060 0.11829 0.15209 0.00 Bilic et al.’s 2.3590 Ours 2.4942 0.39380 –1.18462 –1.07423 0.00 Abe’s 3.6236 Ours 2.4942 0.21259 –0.29483 –0.23524 0.00 Craig et al.’s 2.7592 Ours 2.4942 1.11426 –1.38772 –1.07539 0.00 Arthurs et al.’s 3.7258 160 D. Celik et al.
difficult to use the needle, and dead space within the needle will increase.
Based on a review of the literature, only a few studies have investigated this issue, and these studies reached very different conclusions [9,
19–24]. According to these studies, it was possible to estimate Y using mathematical formulae devel-oped based on age, W, and H values. However, there was no homogeneity between the formulae developed in the different studies. One of the larg-est studies by Chong et al. [20] included 279 pedi-atric oncology patients aged 6 months to 15 years who underwent LP in the lateral position using the L3-L4 or L4-L5 space. In this study, the authors evaluated the relationship between Y and age, W, H, body mass index (BMI), body surface area (BSA), H/W ratio, W/H ratio, and W H. Of these variables, Y showed a strong relation with W, BSA and W/H ratio, whereas it showed no associated with age, ethnic origin and interverte-bral space. In the study by Chong et al. [20], they determined Y using the following formula: Y¼ 10.6 (W/H) þ 0.93. In this formula, the regres-sion coefficient was 0.77. Based on their data,
Chong et al. [20] modified their formula for ease of use as follows; Y¼ 10 (W/H) þ 1.
Bilic et al. [19] described a large series of 195 patients aged 1 day to 19 years, with LPs adminis-tered in the lateral position using the L4-L5 interver-tebral space. According to their analysis of the relationship between Y, H, W, and BSA, Y correlated best with W. They developed the following simple formula for determining Y; Y¼ 1.3 þ 0.07 W. Their most striking result was that the formula could be used in children younger than 3 months. The authors stated that the optimum Y for children younger than 3 months was 1–1.5 cm.
In a similar study, Craig et al. [22] reported in a study which includes 107 children between the ages of 1 day and 16 years underwent LP and they reported that Y correlated best with H. They devel-oped the following formula; Y¼ 0.03 cm H (cm) [22].
In a study of children aged 2–12 years, Rukewe et al. [24] found a positive correlation between Y and W. In a study of pediatric 35 patients, Shenkman et al. [21] determined the distance from the skin to the subarachnoid space. They TABLE 4. Relation with needle depth value in all the models
Ours Chong et al. Bilic et al. Abe et al. Craig et al. Arthurs et al. Needle depth 0.800 0.738 0.783 0.738 0.783 0.783
TABLE 5. PRESS values for all models
Ours Chong et al. Bilic et al. Abe et al. Craig et al. Arthurs et al. PRESS value 36.368 47.379 41.656 323.375 61.779 593.56
administered spinal anesthesia from the L4-L5 intervertebral space using a 25G spinal needle and measured the distance between the skin and sub-arachnoid space. They reported that in premature and former premature infants, this distance corre-lated with PCA and ultrasonographic measure-ments. They developed the following formula; Y (mm) ¼ 13.19 þ 0.0026 W - 0.12 PCA.
Arthurs et al. [23] used ultrasonography (USG) in a study of the estimation and measurement of the distance between skin and the subarachnoid space, measuring the anterior spinal canal depth (ASCD), and posterior spinal canal depth (PSCD). They used the following formula to determine the middle spi-nal caspi-nal depth: (ASCDþ PSCD)/2. To determine the mean spinal canal depth, they used the formula ASCD – PSCD. According to their measurements, both ASCD and PSCD correlated positively with W. ASCD ¼ 1.4 Wþ 5.49 mm (0.60); PSCD ¼ 3.0 Wþ 8.28 mm (0.78); MSCD ¼ 2.2 W þ 6.89 mm (0.76). The most practical formula was accepted as; Y¼ 2 W þ 7 mm. Important characteristics of their study were the inclusion of newborns among the 105 patients and the use of USG as an adju-vant method.
USG has long been used in spinal anesthesia, and its use can prevent the entry to the venous plexus. The use of USG yields more benefit in a seated position than in a lateral position, as the size of the subarachnoid space increases in a seated pos-ition [26]. We did not use USG in this study, as it is not always available in operating theaters and all clinicians are not familiar with its use.
Under the guidance of abdominal computed tomography (CT), Abe et al. [9] measured the dis-tance between skin and the subarachnoid space in 175 patients aged 25 days to 80 years. With the aid of CT, they devised the following formula for calcu-lating Y in LP; Y¼ 1 þ 17 W/H. The wide age range in their study limits the reliability and prac-tical uses of their formula. However, their use of CT, which is a reliable method, was an advantage of their study. In this study, we did not use adjuvant methods, such as USG and CT, due to cost consider-ations and the fact that are not always readily avail-able. In addition, we included only patients from a pediatric age group in our study because the space and tissues display marked variations with age.
Results were obtained for five previously accepted models using data obtained from all indi-viduals. In addition, our model was compared with these five models, and it was found to be different, indicating that a new model has been proposed. When we compared our model with previous mod-els, our model showed the strongest statistical rela-tion with Y. In comparison with the values reported in the other models, those obtained in this model
better estimated. In addition, as shown by the results of statistical tests, the reliability of the model developed herein was better than that of the other models. The models with least reliability were those of Arthurs et al. [23], Chong et al. [20], and Abe et al. [9].
Differences between estimates of the formulae may be the result of various factors, such as differ-ent age ranges, ethnic variations, and differences between clinicians. We believe that the formula developed in this study has high reliability for the specific patient population. The formula developed in this study is based on Hþ W of children aged less than 12 months older than 12 months. Based on a comparison of the reliability of the various formu-lae, we conclude that the use of a formula incorpo-rating Wþ H should be applied in all age groups, as it is more practical.
Our study has some limitations. Values meas-ured at distance L3-L4 may vary. Therefore, new studies can be carried out for that distance and for older children. Individual changes among pediatric patients are too much and should be taken into account.
It should be noted that each child is different and there is a wide variation among individuals, and that this variation should be cautious to esti-mate the LP placement distance.
At the same time, more practical formulas may be needed and we think that new studies should be carried out.
CONCLUSION AND RECOMMENDATIONS In conclusion, it is quite difficult to develop a for-mula representing all patient groups worldwide to calculate the skin-subarachnoid space distance and accordingly choose the size of the needle. However, investigators could develop a formula tailored to the characteristics of patients in their respective regions. Although such formulae will never be 100% accur-ate, they can aid decision-making regarding needle size and increase the success of LP.
DECLARATION OF INTEREST
No potential conflict of interest was reported by the authors.
FUNDING
This study was funded by departmental resources. 162 D. Celik et al.
REFERENCES
[1] K€uc¸€ukali I, Kırbas¸ D. Pratik uygulamada lomber ponk-siyon. D€us¸€unen Adam.1996;9:51–55.
[2] Berg SL, Poplack DG. Treatment of meningeal malig-nancy. Oncologist.1996;1(1 & 2):56–61.
[3] Polaner DM, Drescher J. Pediatric regional anesthesia: what is the current safety record? Paediatr Anaesth.2011; 21(7):737–742. doi:10.1111/j.1460-9592.2010.03499.x. [4] Brown F. History of pediatric regional anesthesia. Paediatr
Anaesth.2012;22(1):3–9. doi:10.1111/j.1460-9592.2011.03636.x. [5] Kachko L, Simhi E, Tzeitlin E, et al. Spinal anesthesia in
neonates and infants—a single-center experience of 505 cases . Paediatr Anaesth. 2007;17(7):647–653. doi:10.1111/ j.1460-9592.2007.02194.x.
[6] Jamil M, Khan Z, Abbasi MM, Ahmad K. Spinal anaes-thesia in paediatric patients undergoing surgery of sub-umbilical region of the body. J Ayub Med Coll Abbottabad.
2011; 23:119–122.
[7] Vandam LD. On the origins of intrathecal anesthesia. Reg Anesth Pain Med.1998;23(4):335–339.
[8] Puncuh F, Lampugnani E, Kokki H. Use of spinal anaes-thesia in paediatric patients: a single centre experience with 1132 cases. Paediatr Anaesth.2004;14(7):564–567. doi: 10.1111/j.1460-9592.2004.01240.x.
[9] Abe KK, Yamamoto LG, Itoman EM, Nakasone TA, Kanayama SK. Lumbar puncture needle length deter-mination. Am J Emerg Med. 2005;23(6):742–746. doi:
10.1016/j.ajem.2005.03.012.
[10] Kokki H. Spinal blocks. Paediatr Anaesth. 2012;22(1): 56–64. doi:10.1111/j.1460-9592.2011.03693.x.
[11] Lopez T, Sanchez FJ, Garzon JC, Muriel C. Spinal anes-thesia in pediatric patients. Minerva Anestesiol.2012;78(1): 78–87. doi:10.1111/j.1460-9592.2011.03769.x.
[12] Apiliogullari S, Duman A, Gok F, Akillioglu I. Spinal needle design and size affect the incidence of postdural puncture headache in children. Paediatr Anaesth. 2010; 20(2):177–182. doi:10.1111/j.1460-9592.2009.03236.x. [13] Apiliogullari S, Duman A, Gok F, Ogun CO, Akillioglu
I. The effects of 45 degree head up tilt on the lumbar puncture success rate in children undergoing spinal anesthesia. Paediatr Anaesth. 2008;18(12):1178–1182. doi: 10.1111/j.1460-9592.2008.02780.x.
[14] Goyal R, Jirtjil K, Baj BB, Singh S, Kumar S. Paediatric spinal anesthesia. Indian J Anesth.2008;52:264–272.
[15] C¸alıs¸kan E, S¸ener M, Koc¸um A, Bozdogan N, Arıbogan A. Pediyatrik hastalarda spinal anestezi deneyimlerimiz our experiences with spinal anesthesia in pediatric patients. Agrı.2011;23:100–106.
[16] Humphreys N, Bays SM, Parry AJ, et al. Spinal anesthe-sia with an indwelling catheter reduces the stress response in pediatric open heart surgery. Anesthesiology.
2005;103(6):1113–1120.
[17] Kachko L, Simhi E, Freud E, et al. Impact of spinal anesthesia for open pyloromyotomy on operating room time. J Pediatr Surg. 2009;44(10):1942–1946. doi:10.1016/ j.jpedsurg.2009.03.038.
[18] Rochette A, Troncin R, Raux O, et al. Clonidine added to bupivacaine in neonatal spinal anesthesia: a prospect-ive comparison in 124 preterm and term infants. Paediatr Anaesth. 2005;15(12):1072–1077. : doi: 10.1111/j.1460-9592.2005.01664.x.
[19] Bilic E, Bilic E, Dadic M, Boban M. Calculating lumbar puncture depth in children. Coll Antropol. 2003;27(2): 623–626.
[20] Chong SY, Chong LA, Ariffin H. Accurate prediction of the needle depth required for successful lumbar punc-ture. Am J Emerg Med. 2010;28(5):603–606. doi:10.1016/ j.ajem.2009.02.006.
[21] Shenkman Z, Rathaus V, Jedeikin R, et al. The distance from the skin to the subarachnoid space can be pre-dicted in premature and former-premature infants. Can J Anaesth.2004; 51:60–62.
[22] Craig F, Stroobant J, Winrow A, Davies H. Depth of insertion of a lumbar puncture needle. Arch Dis Child.
1997;77(5):450doi:10.1136/adc.77.5.450.
[23] Arthurs OJ, Murray M, Zubier M, Tooley J, Kelsall W. Ultrasonographic determination of neonatal spinal canal depth. Arch Dis Child Fetal Neonatal Ed.2008; 93:451–454.
Epub 2008 Feb 19. doi:10.1136/adc.2007.129221.
[24] Rukewe A, Alonge T, Fatiregun A. Spinal anesthesia in children: no longer an anathema! Paediatr Anaesth. 2010; 20(11):1036–1039. doi:10.1111/j.1460-9592.2010.03431.x. [25] Apiliogulları S, G€ok F, Duman A. C¸ocuklarda Spinal
Anestezi: Tek Merkezin 371. Olguluk Deneyimi. T€urk Anest Rean Der Dergisi.2010;38:339–347.
[26] Tsui BC, Suresh S. Ultrasound imaging for regional anesthesia in infants, children, and adolescents: a review of current literature and its application in the practice of neuraxial blocks. Anesthesiology. 2010;112(3):719–728. doi: 10.1097/ALN.0b013e3181c5e03a.
