• Users Online: 385
  • Print this page
  • Email this page

 Table of Contents    
Year : 2012  |  Volume : 56  |  Issue : 3  |  Page : 259-264  

Straight versus flex back: Does it matter in spinal anaesthesia?

1 Department of Anesthesiology, Pain and Perioperative Medicine, ESI Post Graduate Institute of Medical Science and Research, Manicktala, Kolkata, India
2 Department of Anaesthesia, Pain and Hyperbaric Medicine, Royal Adelaide Hospital, North Adelaide SA 5100, Australia
3 Department of Anaesthesiology and Critical Care, B.P. Koirala Institute of Health and Medical Sciences (BPKIHS), Dharan, Nepal
4 Department of Anaesthesiology and Intensive Care, North Eastern Indira Gandhi Regional Institute of Health and Medical Sciences (NEIGRIHMS), Shillong, India

Date of Web Publication20-Jul-2012

Correspondence Address:
Binay Kumar Biswas
Department of Anaesthesiology, Pain and Perioperative Medicine ESI Post Graduate Institute of Medical Science and Research, 54 Bagmari Road, Kolkata 700 054
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/0019-5049.98772

Rights and Permissions

Purpose: It is sometimes difficult for our patients to acquire recommended flexion of their back to perform subarachnoid block upon them. The aim of this study was to find out the degree of procedural success and patient preference when subarachnoid blocks were performed on patients with suboptimal flexion of the back. Methods: Subarachnoid blocks were performed on 160 adult patients from both sexes without any spinal deformity. Procedures were performed with patients in the lateral or seated position with the back either straight or flexed. This approach divided the study population into 4 equal groups: Lateral with back straight (LS) or flexed (LF) and seated with back straight (SS) or flexed (SF). The primary endpoint was correct needle placement. Numbers of attempts, needle redirections and patients' preferred posture were determined to compare the outcome in different groups. Results: For both positions, the overall success rates were 95% and 100% in patients who had straight or flexed back, respectively (P=0.81). In the lateral position, significantly more patients of the LF group (40) than those of the LS group (32) had successful placement of spinal needle at first attempt (P=0.03). Altogether, 34 and 21 patients in the SS and SF groups, respectively, required cephalad redirections of the needle (P=0.003). Most patients preferred the straight back position (69.7-88%). Conclusion: With a higher preference by patients for the straight back posture, the overall success rate of correct spinal needle placement was comparable among the groups who had their back placed in a straight or in a flexed posture for subarachnoid block.

Keywords: Anaesthesia neuraxial, attitude, position, spinal

How to cite this article:
Biswas BK, Agarwal B, Bhattarai B, Dey S, Bhattacharyya P. Straight versus flex back: Does it matter in spinal anaesthesia?. Indian J Anaesth 2012;56:259-64

How to cite this URL:
Biswas BK, Agarwal B, Bhattarai B, Dey S, Bhattacharyya P. Straight versus flex back: Does it matter in spinal anaesthesia?. Indian J Anaesth [serial online] 2012 [cited 2020 Nov 27];56:259-64. Available from: https://www.ijaweb.org/text.asp?2012/56/3/259/98772

   Introduction Top

Be it a sitting or a lateral position, patients are placed with their back flexed to perform subarachnoid block for spinal anaesthesia. [1],[2] Flexed back for subarachnoid block is considered mandatory because of widening of the interspinous space in this state as well as paucity of literature regarding patients' preference for any particular posture. [3],[4] It may be uncomfortable for some patients to assume a flexed posture. [5],[6],[7] Straight back is a suboptimal posture to perform spinal block. As there is no study carried out to compare the success of lumbar puncture associated with these postures, it is difficult to predict the outcome of each. In this prospective, randomized and observational study involving healthy adults, we have worked on the above issues to determine the patients' preferred posture and success of subarachnoid block performed in an optimal flexed posture and in a posture that is considered suboptimal flexion for the procedure.

   Methods Top

After obtaining approval from the Institute Ethical Committee and written informed consent from patients, 160 adult patients of the age group of 18-60 years of either sex belonging to ASA-PS 1 and 2, scheduled for elective surgery under spinal anaesthesia were included in the study. Patients with anatomical spinal deformity, body mass index (BMI)>30 kg/m 2 , previous back surgery, non-palpable spinous process, neurological disease, coagulation disorders, local infection and unstable haemodynamics were excluded from the study. Spinal block was performed either in the lateral or in the sitting position. Patients' were divided into four groups of 40 patients each based on a computer-generated randomization system: Lateral position with straight spine (LS), lateral position with flex spine (LF), sitting position with straight spine (SS) and sitting position with flex spine (SF). During the pre-anaesthetic meeting on the evening before surgery, patients were explained about the proposed postures for the procedure according to the randomization table. Moreover, to assess patients' preference, each one of them was explained that they would be placed in both flex and straight postures before putting them into the final performing position. The time to be kept in non-performing posture (straight for flex group and vice versa) would be 2 min. They were asked to tell us their preferred posture with reason (s) for it just after completion of the procedure.

All patients were premedicated with oral lorazepam 2 mg the night before and on the morning. In the operation theatre, monitors (electrocardiogram, pulse oximetry and non-invasive blood pressure) were attached and patients were prepared for subarachnoid block. Patients of the LF and LS groups were placed in a lateral position at the edge of the table with one pillow under the head. The patients of the LF group were initially placed in a straight state, keeping the hip joints at a right angle with the back. This brought both iliac crests into the same vertical plane without any forward and backward tilt of the lower back. Scapulae were also kept into the same vertical plane with help from alignments of shoulder joints while placing the upper hand over the head with the elbow flexed. These maneuvers turned the upper back straight without any forward and backward tilt. The head was supported with a pillow so as to keep the neck in a comfortable position. This whole procedure provided us a laterally lying patient without any forced flexion of the back. After 2 min, patients were asked to acquire flex posture by bending the hip joints and flexing the neck and knee towards each other as far as possible (forehead to knee position). An anaesthesia assistant helped each patient to obtain and maintain the best possible flexed posture by holding the patients at their occipital region and knees. [1],[2] Block was performed in the lateral position. Similarly, patients of the LS group were first placed in a flex state and then were put in the final straight posture for spinal block following the above-mentioned methods.

Patients of the SF and SS groups sat on the table with their legs hanging from the edge of the table with support of a stool under their feet. The height of the stool was adjusted to achieve a right angle between the thigh and the back (hip joints of the SF group were initially positioned with a straight back). We considered the sitting straight back posture (SS) as the position without any voluntarily achieved flexion of the back. To achieve a straight back posture, patients placed both of their hands over their respective thighs with the elbows flexed. Both arms remained at a respective mid-axillary plane thus avoiding a forward or backward tilt of the scapulae. Sitting posture-induced normal flexion of back, if any, was accepted; however, no active or assisted flexion of back was performed. To achieve a sitting flex posture (SF), patients were asked to bend forward neck and arch out back maximally. Arms were rested over a pillow kept on the lap. This prevents the patient from slumping to either side. [2] An assistant standing in front supported the patient to maintain the posture. Patients of the SS group were initially placed with their back flexed followed by straight posture as per the methods described above.

Under aseptic conditions and following standard techniques of spinal anaesthesia, the second author (having more than 5 years of experience in neuraxial block) performed the blocks with a 25 G, 3.5-inch-long Quincke spinal needle (B. Braun, Melsungen AG, Germany). The author sat on a stool and the height of both the operating table and the author's stool was adjusted so as to keep the interspinous space at the level of the performer's eye. Taking iliac crest as the marker, the performer decided his first and the second preferred intervertebral space out of L2-L3, L3-L4 and L4-L5.

In the first attempt, the needle was introduced at the first preferred space with a slight cephalad direction, and its placement into the subarachnoid space was judged by tactile sensation. After withdrawing the stylet, appearance of cerebrospinal fluid (CSF) in the hub of the needle was looked for. Aspiration of free flow of CSF was considered as successful needle insertion and spinal anaesthesia was continued with administration of 0.5% heavy bupivacaine.

From perceived tactile sensation, when it was felt that the needle placement was in correct position but there was either no CSF in the needle hub or there was scanty CSF with poor flow, the following adjustments were performed for proper placement of needle. The needle was rotated clockwise 90° and waited for 5 s; sequence of rotation continued for another three-quadrant rotation of 90°, each amounting to 360°, with an interval of 5 s between each rotation. Despite this, if there was absence of CSF or its free flow, the needle was further advanced by approximately 2 mm. Failure to achieve CSF or its free flow even after this maneuver was followed by withdrawal of the needle by 2 mm. Absence of CSF or its free flow after all these maneuvers was considered as failed first attempt.

Following failure of the first attempt, the needle was removed completely; reassessment of the midline was performed and the needle was reintroduced in the same first preferred space (second attempt). After its placement in the subarachnoid space, as judged by tactile sensation, free flow of CSF was looked for and taken as successful placement. In the absence of CSF or its free flow, the above-mentioned maneuvers were repeated either to have a successful needle placement or to consider it as to be failed second attempt.

In case of failed second attempt, the third and the final attempt were considered in second preferred space (either one space above or below the first preferred space) involving all above procedures to get a free flow CSF. However, in case of failed third attempt, repeat midline assessment for second preferred space was not undertaken for another attempt (as was done for first preferred space) and spinal procedure was abandoned in favour of general anaesthesia.

When bone was encountered during any of the above-mentioned attempts, the needle was withdrawn just below the skin level followed by reinsertion with a more cephalad angulation. Persistent contact (more than two-times) with bone at the same depth was thought to be deviation off-midline and the needle was withdrawn. After confirming midline, needles were reintroduced in the same space, and the above-mentioned redirections (if needed) were followed in cases of bony contacts. In each attempt, the number of redirections of the needle was limited to two. More than two contacts with the bone even after confirming the midline was considered as failure of that particular attempt. Redirection of needle or its reintroduction after confirmation of midline was taken as a part of the individual attempt of spinal block (first, second or third as described before) to count the number of redirections required in each attempt. Shooting pain experienced by the patients or jerky movements of the limbs during the procedure was noted. Thus, considering all the above-mentioned manipulations, each attempt was considered as a failed attempt if: (i) there was no CSF in the hub, (ii) CSF was present but without free flow and (iii) when the needle could not be negotiated because of repeated bony contacts. A patient with traumatic tap in straight posture was placed in the respective flexed posture and another attempt of needle placement was performed in the next best space. For patients in the flexed group, after the first traumatic tap, the procedure was attempted in the next best space. The procedure was abandoned if there was trauma even in the second space. Patients having procedural trauma were followed-up for 24 h.

Following administration of local anaesthetic, patients were positioned supine and adequacy of analgesia was confirmed after 10 min of spinal block to proceed for surgery. Patients with insufficient analgesia received general anaesthesia. The following parameters were recorded for data analysis: Overall success, success without any manipulation of needles, attempts required, number of patients requiring manipulation of needle and the type of manipulations. Preference of the patients' position was assessed by asking them which position they had preferred more. Direct questions were also asked as to why had they preferred such a position and why not the other? They were not provided any affirmative or negative suggestions (like whether you had pain in limbs, abdomen, neck, breathing difficulties, chest compression, etc.) to get the answers from them. Other post-operative complications like headache and backache were noted by the performer after the operation.

Power analysis based on a high overall success rate of subarachnoid block with the 25 G Quincke spinal needle (97%) required 15 patients in each group to complete the study with an α and ß error of 5 and10%, respectively. [8] Subarachnoid block is a routine practice in our institution without involvement of extra cost, and also because of the availability of patients, where we involved 40 patients in each group to complete the study. Normally distributed numerical variables among the four groups were analyzed with one-way analysis of variance. Metric variables between the two groups were analyzed with unpaired Student's t-test. Proportions including sex distributions were analyzed with Chi-square or Fisher's exact tests with Yates' correction. Reasons behind patients' preferences or refusals of any particular position have been described in percentage only because no predefined ordinal or categorical data were used for their evaluation. In all calculations, P<0.05 was the level of significance.

   Results Top

Demographic parameters (age, sex and BMI) were comparable among the groups [Table 1].
Table 1: Demographic data

Click here to view

A spinal needle could not be placed correctly in two patients each in groups LS and SS, even with three attempts. Therefore, there were 40 patients each in the LF and SF groups and 38 patients in the LS and SS groups for further analyses. Overall success rates of subarachnoid needle placement were 95% (38), 100% (40), 95% (38) and 100% (40) for the LS, LF, SS and SF group, respectively. [Table 2] shows the comparative values of various other performance parameters among the groups. Patients in the LF group did not require redirection of needles. No patients in any of the groups required needle redirection because of lack or inadequate flow of CSF. All redirections were because of repeated bony contacts. Redirections required at first attempt were noted in 13, 28 and 17 patients among the LS, SS and SF groups, respectively. The rest of the patients needed them at the second attempt. Five, three and two patients in the SS, LS and SF groups, respectively, had pain suggestive of nerve root irritation during the procedure without any development of neurodeficit. One patient each in the LF and SF groups had headache persisting for 48 h.
Table 2: Successful placement of spinal needle

Click here to view

Most of the patients (69.7-88%, depending on the group) preferred the straight posture. Except 8% of the patients of the LS group, no patients of any other group expressed their preference for flexed posture. However, few patients (16-26%, depending on the group) were unable to decide their preferred posture and indicated their likings for either of the postures.

In the LS, LF, SS and SF groups, respectively, 67.5, 45, 47.5 and 45% of the patients reported that they had preferred the particular posture without any specific reasons; however, 20, 52.5, 52.5 and 42.5% of the patients in the same groups expressed that they simply were more comfortable in their preferred postures. In the LS group, 12.5% of the patients preferred the straight posture as they had felt pain over the abdomen in the flex posture. Because of discomfort over the neck in the flex posture, 12.5% of the patients in the SF group preferred the straight posture. In the LF group, 2.5% of the patients mentioned that they had discomfort both at the knee joints and over the abdomen in the flex posture. However, few patients of the LS group (8%) liked the flex posture as they had the feeling of being more secure with assistance from the attendants who were holding them.

   Discussion Top

Success of spinal block with the 25 G spinal Quincke spinal needle may be as high as 97% in adults and 100% in the paediatric population. [8],[9] Overall success of spinal block in our patients is 100% in patients with flexed posture and 95% in patients with straight back posture. First-attempt success rates (84.2-100%) in our study are superior to the reported success rate of 61.5% in the adult population. [7] Higher success rates may be because of strict exclusion criteria adopted in our study.

In our study, the overall success rate of subarachnoid block in patients with flexed and straight spine of the two positions (sitting and lateral) were comparable. However, first-attempt success rate is significantly lower in patients who had straight (LS) back than in those having flexed back (LF) in lateral position. Although overall requirements of needle redirections were comparable between these two groups, more patients of the straight back posture (LS) required midline reassessment resulting in a decreased success rate at first attempt and increase in the number of second attempts. Bony contacts could be negotiated easily by cephalad redirection of needles. In the lateral position, midline, as evidenced by longitudinal depression of skin, often does not correspond with the underlying vertebral column. This effect may be more pronounced if the back is not flexed. This could be the reason behind the difficulties to correctly identify midline in six of our patients of the LS group.

Success rate at first attempt is comparable between patients who had straight or flex back in sitting position (SS and SF). In sitting position, bony midline remains more accurately beneath the midline as evidenced on skin; this might have been the reason for the comparable success rate at first attempt between the groups.

In the sitting state, significantly more number of patients with straight back position (SS) required needle redirections than patients with flex back (SF). Number of patients requiring needle redirection is comparable between the LS and LF groups. It is difficult to explain why patients of the SS group required more needle redirections. Cephalad direction that was applied at needle entry points was, most probably, not adequate for proper alignment of the needle in this position to reach subarachnoid space without touching the lamina. However, requirement of more numbers of needle redirections neither did influence the number of second attempts nor the overall success rate. As sitting position maintains midline better, there was no difference in requirement of second attempt between the groups (SS and SF) in the sitting position.

Most of the patients preferred straight postures. As opined by patients, flex posture was often associated with discomfort to neck and pain over abdomen as well as in knee joints. There were few patients in every group who could not decide their preferred postures. In a crossover manner, we kept the patient for 2 min in a non-performing state. In the performing posture, patients remain for a longer period (although not measured, yet it is correct to say that spinal block takes more than 2 min) than the non-performing posture. It may be perceived that an achieved flex posture of longer duration is more complex and uncomfortable than the straight posture. Because of this, probably most patients preferred the straight posture. When flex posture was the non-performing state, the patient might not have fully perceived the discomfort associated with this posture (as they had this posture for 2 min only). This might have been the reason for a large number of patients to remain indifferent about the preferred posture.

Our study is the first of its kind showing different aspects of the technical parameters associated with spinal block performed on patients with suboptimal postures. This study documents various such parameters and shows that success is similar to that of conventional posture with increased occurrence of negotiable technical difficulties. The results may not be comparable for spinal blocks performed with spinal needles requiring introducers. Moreover, results of our study come from an experienced performer; thereby, results in similar situations may vary for beginners in the field of regional anaesthesia. Novices should start training with the gold standard posture, but experienced operators may not insist on unnecessary flex posture in easy patients.

Our idea was not to challenge the superiority of the classical posture. Rather, it was to find out the feasibility of spinal block in patients with suboptimal postures. As we included patients with BMI<30, results may differ in patients with higher BMI. Further studies involving patients from various racial origins may provide necessary technical inputs regarding spinal block in a suboptimal position among these groups. Moreover, future multicentric studies, possibly focusing on various sub-populations (viz. limb trauma, arthritis) are warranted to obtain specific data relevant to those populations that cannot have their back adequately flexed for spinal anaesthesia. An open questionnaire-based evaluation of preference has its own limitations; yet, we tried to avoid such limitations as much as possible by avoiding affirmative or negative questions.

   Conclusion Top

Lumbar puncture can be performed with equal ease in a seated patient regardless of the posture of the back being in a flex or in a straight posture, provided the landmarks are clear. Most of the patients prefer straight back to the flexed back posture because of the discomfort associated with the latter state.

   References Top

1.Brown DL. Spinal epidural and caudal anesthesia, Miller Anesthesia, In: Miller RD, Editor, 6 th ed. Philadelphia (US): Elsevier Churchill Livingstone; 2005.p. 1661-4  Back to cited text no. 1
2.Bernards CM. Epidural and spinal anesthesia, Clinical Anesthesia, In: Barash PG, Cullen BF, Stoelting RK, Editors. 5 th ed. Philadelphia: Lipincott-Raven Publishers; 2006.p. 696-8.  Back to cited text no. 2
3.Fisher A, Lupu L, Gurevitz B, Brill E, Margolin E, Hertzanu Y. Hip flexion and lumbar puncture: A radiological study. Anaesthesia 2001;56:262-6.  Back to cited text no. 3
4.Sandoval M, Shestak W, Sturmann K, Hsu C. Optimal patient position for lumbar puncture, measured by ultrasonography. Emerg Radiol 2004;4:179-81.  Back to cited text no. 4
5.Platt F, Hall M. Hip flexion for lumbar puncture-not easy for obstetric practice. Anesthesia 2001;56:1023.  Back to cited text no. 5
6.Schelew BL, Vaghadia H. Ankylosing Spodylitis and neuroaxial anesthesia: A 10 year review. Can J Anaesth 1996;43:55-8.  Back to cited text no. 6
7.DeOliveira GR, Gomes HP, da Fonseca MH, Hoffman JC, Pederneiras SG, Garcia JH. Predictors of successful neuroaxial block; a prospective study. Eur J Anaesthesiol 2002;19:447-51.  Back to cited text no. 7
8.Tarkkila P, Huhtala J, SalminenU. Difficulties in spinal needle use: Insertion characteristics and failure rates associated with 25-, 27-and 29-gauge Quincke-type spinal needles. Anaesthesia 1994;49:723-5.  Back to cited text no. 8
9.Kokki H, Hendolin H. Comparison of 25 G and 29G Quincke spinal needles in paediatric day case surgery: A prospective randomized study of the puncture characteristics, success rate and post operative complaints. Paediatr Anaesth 1996;6:115-9.  Back to cited text no. 9


  [Table 1], [Table 2]

This article has been cited by
1 Comparing Three Different Modified Sitting Positions for Ease of Spinal Needle Insertion in Patients Undergoing Spinal Anesthesia
Sussan Soltani Mohammadi,Mohammadreza Piri,Alireza Khajehnasiri
Anesthesiology and Pain Medicine. 2017; In Press(In Press)
[Pubmed] | [DOI]
2 Optimal Point of Insertion and Needle Angle in Neuraxial Blockade Using a Midline Approach
Mark Vogt,Dennis J. van Gerwen,Wouter Lubbers,John J. van den Dobbelsteen,Martin Hagenaars
Regional Anesthesia and Pain Medicine. 2017; 42(5): 600
[Pubmed] | [DOI]
3 Comparison of Successful Spinal Needle Placement Between Crossed-Leg Sitting Position and Traditional Sitting Position in Patients Undergoing Urology Surgery
Sidharta K. Manggala,Aida R. Tantri,Darto Satoto
Anesthesiology and Pain Medicine. 2016; In press(In press)
[Pubmed] | [DOI]
4 Comparing the Squatting Position and Traditional Sitting Position for Ease of Spinal Needle Placement: A Randomized Clinical Trial
Sussan Soltani Mohammadi,Marzieh Hassani,Seyed Mojtaba Marashi
Anesthesiology and Pain Medicine. 2014; 4(2)
[Pubmed] | [DOI]
5 Patient position for spinal anaesthesia: Flexed-back versus straight-back
Prakash, S.
Indian Journal of Anaesthesia. 2013; 57(1): 95
6 Ultrasound-guided intrathecal anesthesia: Does scanning help?
Sherif A. Abdelhamid,Magdy A. Mansour
Egyptian Journal of Anaesthesia. 2013; 29(4): 389
[Pubmed] | [DOI]


    Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
    Access Statistics
    Email Alert *
    Add to My List *
* Registration required (free)  

  In this article
    Article Tables

 Article Access Statistics
    PDF Downloaded648    
    Comments [Add]    
    Cited by others 6    

Recommend this journal